{"data":{"skill":{"slug":"prime-skills-flux-2-klein","name":"flux-2-klein","icon":"📦","repo":"https://github.com/prime-skills/runcomfy-agent-skills/tree/fca19ae084c2ae4667f5c8affbb99da5103d5832/flux-2-klein","status":"approved","author":"prime-skills","authorVersion":null,"skillstoreRevision":1,"license":"MIT"},"licenseCorrection":null,"audit":{"id":"ade34d8c-3664-4213-aacf-6ec3961c98bd","skill_id":"9937e27e-e5ad-416e-86db-f0c3466eb298","version":1,"content_hash":"v3:e4fe64cde5e2867f53e31488fa673df18e2e8130:f808637da840042e899b5186ba35d5aa24bfa5585151e48e822ca67c030b350a:df7bd662cf42551a752f518066f9856b38c7bdd3c2b06204b3a7ead609c59b4b:736b696c6c732f7072696d652d736b696c6c732f666c75782d322d6b6c65696e:ecdea592cecce1867fab1223e1c87be4","risk_level":"medium","is_blocked":false,"safe_to_publish":true,"analysis_status":"ok","agent_auto_install_policy":"allowed","manual_install_policy":"allowed","summary":"All 57 static findings are false positives involving Markdown formatting, expected CLI operations, documented service URLs, API abbreviations, and credential storage. No evidence found of malicious intent, credential theft, or audit-directed prompt injection. One semantic concern remains: shell examples omit prompt escaping while asserting injection safety; the external CLI implementation was not available for verification.","remediation":[{"issue":"Prompt placeholders appear inside single-quoted shell arguments, but the safety statement overlooks quoting performed by the invoking shell.","severity":"medium","suggestion":"Use JSON serialization and argument arrays with shell execution disabled. If shell commands remain, document safe argument escaping and test apostrophes and metacharacters."}],"risk_factor_evidence":[{"factor":"external_commands","evidence":[{"file":"SKILL.md","line_end":13,"line_start":13},{"file":"SKILL.md","line_end":14,"line_start":14},{"file":"SKILL.md","line_end":29,"line_start":27},{"file":"SKILL.md","line_end":49,"line_start":29},{"file":"SKILL.md","line_end":50,"line_start":49},{"file":"SKILL.md","line_end":51,"line_start":50},{"file":"SKILL.md","line_end":51,"line_start":51},{"file":"SKILL.md","line_end":63,"line_start":57},{"file":"SKILL.md","line_end":64,"line_start":63},{"file":"SKILL.md","line_end":65,"line_start":64},{"file":"SKILL.md","line_end":66,"line_start":65},{"file":"SKILL.md","line_end":66,"line_start":66},{"file":"SKILL.md","line_end":72,"line_start":68},{"file":"SKILL.md","line_end":76,"line_start":72},{"file":"SKILL.md","line_end":82,"line_start":76},{"file":"SKILL.md","line_end":86,"line_start":82},{"file":"SKILL.md","line_end":90,"line_start":86},{"file":"SKILL.md","line_end":99,"line_start":90},{"file":"SKILL.md","line_end":103,"line_start":99},{"file":"SKILL.md","line_end":107,"line_start":103},{"file":"SKILL.md","line_end":109,"line_start":107},{"file":"SKILL.md","line_end":109,"line_start":109},{"file":"SKILL.md","line_end":116,"line_start":113},{"file":"SKILL.md","line_end":122,"line_start":116},{"file":"SKILL.md","line_end":135,"line_start":122},{"file":"SKILL.md","line_end":158,"line_start":135},{"file":"SKILL.md","line_end":163,"line_start":158},{"file":"SKILL.md","line_end":167,"line_start":163},{"file":"SKILL.md","line_end":171,"line_start":167},{"file":"SKILL.md","line_end":175,"line_start":171},{"file":"SKILL.md","line_end":179,"line_start":175},{"file":"SKILL.md","line_end":191,"line_start":179},{"file":"SKILL.md","line_end":197,"line_start":191},{"file":"SKILL.md","line_end":206,"line_start":197},{"file":"SKILL.md","line_end":207,"line_start":206},{"file":"SKILL.md","line_end":208,"line_start":207},{"file":"SKILL.md","line_end":208,"line_start":208},{"file":"SKILL.md","line_end":209,"line_start":209},{"file":"SKILL.md","line_end":210,"line_start":210},{"file":"SKILL.md","line_end":219,"line_start":219},{"file":"SKILL.md","line_end":222,"line_start":220},{"file":"SKILL.md","line_end":222,"line_start":222}]},{"factor":"network","evidence":[{"file":"SKILL.md","line_end":17,"line_start":17},{"file":"SKILL.md","line_end":23,"line_start":23},{"file":"SKILL.md","line_end":76,"line_start":76},{"file":"SKILL.md","line_end":202,"line_start":202},{"file":"SKILL.md","line_end":207,"line_start":207}]},{"factor":"filesystem","evidence":[{"file":"SKILL.md","line_end":208,"line_start":208},{"file":"SKILL.md","line_end":209,"line_start":209},{"file":"SKILL.md","line_end":210,"line_start":210},{"file":"SKILL.md","line_end":219,"line_start":219},{"file":"SKILL.md","line_end":219,"line_start":219}]}],"critical_findings":[],"high_findings":[],"medium_findings":[{"title":"Shell Quoting Safety Is Incorrectly Assumed","locations":[{"file":"SKILL.md","line_end":98,"line_start":82},{"file":"SKILL.md","line_end":106,"line_start":103},{"file":"SKILL.md","line_end":220,"line_start":220}],"confidence":0.88,"description":"Examples place <user prompt> inside single-quoted shell JSON without escaping guidance. Literal substitution of an apostrophe can terminate the shell argument before CLI processing. The statement \"No shell injection surface from prompt content\" overlooks this boundary and can encourage unsafe command construction.","review_kind":"security","source_category":"semantic","source_severity":"medium","confidence_reasoning":"The examples and absolute safety claim are explicit. Exploitation depends on literal prompt substitution; no executable interpolation implementation was supplied."}],"low_findings":[],"dangerous_patterns":[],"files_scanned":1,"total_lines":224,"audit_model":"codex","audited_at":"2026-11-14T14:43:05.664+00:00","created_at":"2026-11-14T18:16:35.031602+00:00","static_findings":[{"id":"external_commands:SKILL.md:13:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"`runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or","category":"external_commands","line_end":13,"severity":"medium","line_start":13},{"id":"external_commands:SKILL.md:14:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"`/4b/`) through the local RunComfy CLI. Triggers on \"flux 2 klein\",","category":"external_commands","line_end":14,"severity":"medium","line_start":14},{"id":"external_commands:SKILL.md:27:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```bash","category":"external_commands","line_end":29,"severity":"medium","line_start":27},{"id":"external_commands:SKILL.md:29:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":49,"severity":"medium","line_start":29},{"id":"external_commands:SKILL.md:49:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"1. **RunComfy CLI** — `npm i -g @runcomfy/cli`","category":"external_commands","line_end":50,"severity":"medium","line_start":49},{"id":"external_commands:SKILL.md:50:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.","category":"external_commands","line_end":51,"severity":"medium","line_start":50},{"id":"external_commands:SKILL.md:51:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"3. **CI / containers** — set `RUNCOMFY_TOKEN=<token>` instead of `runcomfy login`.","category":"external_commands","line_end":51,"severity":"medium","line_start":51},{"id":"external_commands:SKILL.md:57:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"### `blackforestlabs/flux-2-klein/9b/text-to-image`","category":"external_commands","line_end":63,"severity":"medium","line_start":57},{"id":"external_commands:SKILL.md:63:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"| `prompt` | string | yes | — | Up to ~512 tokens. Longer degrades. |","category":"external_commands","line_end":64,"severity":"medium","line_start":63},{"id":"external_commands:SKILL.md:64:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"| `steps` | int | no | 25 | 4–50. **Step-distilled architecture** — 4–8 enough for concepting; ~25 f","category":"external_commands","line_end":65,"severity":"medium","line_start":64},{"id":"external_commands:SKILL.md:65:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"| `width` | int | no | 1024 | 512–1536 typical. **Aspect ratio capped at 16:9**, max ~2K total. |","category":"external_commands","line_end":66,"severity":"medium","line_start":65},{"id":"external_commands:SKILL.md:66:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"| `height` | int | no | 1024 | Match `width`'s aspect intent. |","category":"external_commands","line_end":66,"severity":"medium","line_start":66},{"id":"external_commands:SKILL.md:68:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"### `blackforestlabs/flux-2-klein/4b/text-to-image`","category":"external_commands","line_end":72,"severity":"medium","line_start":68},{"id":"external_commands:SKILL.md:72:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"Same field set as 9B. Default `steps` is effectively 4 — the variant is built for that step count.","category":"external_commands","line_end":76,"severity":"medium","line_start":72},{"id":"external_commands:SKILL.md:76:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"Up to **4 simultaneous reference images** are supported on the same endpoint for style transfer / gu","category":"external_commands","line_end":82,"severity":"medium","line_start":76},{"id":"external_commands:SKILL.md:82:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```bash","category":"external_commands","line_end":86,"severity":"medium","line_start":82},{"id":"external_commands:SKILL.md:86:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":90,"severity":"medium","line_start":86},{"id":"external_commands:SKILL.md:90:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```bash","category":"external_commands","line_end":99,"severity":"medium","line_start":90},{"id":"external_commands:SKILL.md:99:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":103,"severity":"medium","line_start":99},{"id":"external_commands:SKILL.md:103:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```bash","category":"external_commands","line_end":107,"severity":"medium","line_start":103},{"id":"external_commands:SKILL.md:107:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":109,"severity":"medium","line_start":107},{"id":"external_commands:SKILL.md:109:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"The CLI submits, polls every 2s until terminal, then downloads any `*.runcomfy.net` / `*.runcomfy.co","category":"external_commands","line_end":109,"severity":"medium","line_start":109},{"id":"external_commands:SKILL.md:113:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```bash","category":"external_commands","line_end":116,"severity":"medium","line_start":113},{"id":"external_commands:SKILL.md:116:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":122,"severity":"medium","line_start":116},{"id":"external_commands:SKILL.md:122:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"**Subject-first declarative grammar.** The structure Flux 2 Klein was trained on is *\"Subject + acti","category":"external_commands","line_end":135,"severity":"medium","line_start":122},{"id":"external_commands:SKILL.md:135:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"**For text rendering** (Klein has the 8B Qwen3 embedder, decent but not GPT Image 2 territory): add ","category":"external_commands","line_end":158,"severity":"medium","line_start":135},{"id":"external_commands:SKILL.md:158:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":163,"severity":"medium","line_start":158},{"id":"external_commands:SKILL.md:163:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":167,"severity":"medium","line_start":163},{"id":"external_commands:SKILL.md:167:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":171,"severity":"medium","line_start":167},{"id":"external_commands:SKILL.md:171:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":175,"severity":"medium","line_start":171},{"id":"external_commands:SKILL.md:175:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":179,"severity":"medium","line_start":175},{"id":"external_commands:SKILL.md:179:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"```","category":"external_commands","line_end":191,"severity":"medium","line_start":179},{"id":"external_commands:SKILL.md:191:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"The `runcomfy` CLI uses sysexits-style codes:","category":"external_commands","line_end":197,"severity":"medium","line_start":191},{"id":"external_commands:SKILL.md:197:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"| 65 | bad input JSON / schema mismatch (e.g. `width: 4096` would 422) |","category":"external_commands","line_end":206,"severity":"medium","line_start":197},{"id":"external_commands:SKILL.md:206:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"1. The skill invokes `runcomfy run blackforestlabs/flux-2-klein/<variant>/text-to-image` with a JSON","category":"external_commands","line_end":207,"severity":"medium","line_start":206},{"id":"external_commands:SKILL.md:207:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"2. The CLI POSTs to `https://model-api.runcomfy.net/v1/models/blackforestlabs/flux-2-klein/<variant>","category":"external_commands","line_end":208,"severity":"medium","line_start":207},{"id":"external_commands:SKILL.md:208:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"3. The Model API returns a `request_id`; the CLI polls `GET .../requests/<id>/status` every 2 second","category":"external_commands","line_end":208,"severity":"medium","line_start":208},{"id":"external_commands:SKILL.md:209:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"4. On terminal status, the CLI fetches `GET .../requests/<id>/result` and downloads any URL whose ho","category":"external_commands","line_end":209,"severity":"medium","line_start":209},{"id":"external_commands:SKILL.md:210:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"5. `Ctrl-C` while polling sends `POST .../requests/<id>/cancel` so you don't get billed for GPU you ","category":"external_commands","line_end":210,"severity":"medium","line_start":210},{"id":"external_commands:SKILL.md:219:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m","category":"external_commands","line_end":219,"severity":"medium","line_start":219},{"id":"external_commands:SKILL.md:220:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"- **Input boundary**: the user prompt is passed as a JSON string to the CLI via `--input`. The CLI d","category":"external_commands","line_end":222,"severity":"medium","line_start":220},{"id":"external_commands:SKILL.md:222:ruby-shell-backtick-execution","file":"SKILL.md","pattern":"Ruby/shell backtick execution","snippet":"- **Outbound endpoints**: only `model-api.runcomfy.net` (request submission) and `*.runcomfy.net` / ","category":"external_commands","line_end":222,"severity":"medium","line_start":222},{"id":"network:SKILL.md:17:hardcoded-url","file":"SKILL.md","pattern":"Hardcoded URL","snippet":"homepage: https://www.runcomfy.com","category":"network","line_end":17,"severity":"low","line_start":17},{"id":"network:SKILL.md:23:hardcoded-url","file":"SKILL.md","pattern":"Hardcoded URL","snippet":"[runcomfy.com](https://www.runcomfy.com/?utm_source=skills.sh&utm_medium=skill&utm_campaign=flux-2-k","category":"network","line_end":23,"severity":"low","line_start":23},{"id":"network:SKILL.md:76:hardcoded-url","file":"SKILL.md","pattern":"Hardcoded URL","snippet":"Up to **4 simultaneous reference images** are supported on the same endpoint for style transfer / gu","category":"network","line_end":76,"severity":"low","line_start":76},{"id":"network:SKILL.md:202:hardcoded-url","file":"SKILL.md","pattern":"Hardcoded URL","snippet":"Full reference: [docs.runcomfy.com/cli/troubleshooting](https://docs.runcomfy.com/cli/troubleshootin","category":"network","line_end":202,"severity":"low","line_start":202},{"id":"network:SKILL.md:207:hardcoded-url","file":"SKILL.md","pattern":"Hardcoded URL","snippet":"2. The CLI POSTs to `https://model-api.runcomfy.net/v1/models/blackforestlabs/flux-2-klein/<variant>","category":"network","line_end":207,"severity":"low","line_start":207},{"id":"filesystem:SKILL.md:208:path-traversal-sequence","file":"SKILL.md","pattern":"Path traversal sequence","snippet":"3. The Model API returns a `request_id`; the CLI polls `GET .../requests/<id>/status` every 2 second","category":"filesystem","line_end":208,"severity":"high","line_start":208},{"id":"filesystem:SKILL.md:209:path-traversal-sequence","file":"SKILL.md","pattern":"Path traversal sequence","snippet":"4. On terminal status, the CLI fetches `GET .../requests/<id>/result` and downloads any URL whose ho","category":"filesystem","line_end":209,"severity":"high","line_start":209},{"id":"filesystem:SKILL.md:210:path-traversal-sequence","file":"SKILL.md","pattern":"Path traversal sequence","snippet":"5. `Ctrl-C` while polling sends `POST .../requests/<id>/cancel` so you don't get billed for GPU you ","category":"filesystem","line_end":210,"severity":"high","line_start":210},{"id":"filesystem:SKILL.md:219:hidden-file-in-home-directory","file":"SKILL.md","pattern":"Hidden file in home directory","snippet":"- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m","category":"filesystem","line_end":219,"severity":"high","line_start":219},{"id":"filesystem:SKILL.md:219:hidden-file-access","file":"SKILL.md","pattern":"Hidden file access","snippet":"- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m","category":"filesystem","line_end":219,"severity":"medium","line_start":219},{"id":"blocker:SKILL.md:65:at-command-scheduling","file":"SKILL.md","pattern":"at command scheduling","snippet":"| `width` | int | no | 1024 | 512–1536 typical. **Aspect ratio capped at 16:9**, max ~2K total. |","category":"blocker","line_end":65,"severity":"medium","line_start":65},{"id":"blocker:SKILL.md:33:system-reconnaissance","file":"SKILL.md","pattern":"System reconnaissance","snippet":"Flux 2 Klein's distinct strength is **latency-first creative iteration**: sub-second feedback enable","category":"blocker","line_end":33,"severity":"low","line_start":33},{"id":"blocker:SKILL.md:45:system-reconnaissance","file":"SKILL.md","pattern":"System reconnaissance","snippet":"If the user said \"Flux 2 Klein\" / \"BFL Klein\" / \"flux klein\" explicitly, route here regardless. If t","category":"blocker","line_end":45,"severity":"low","line_start":45},{"id":"blocker:SKILL.md:115:system-reconnaissance","file":"SKILL.md","pattern":"System reconnaissance","snippet":"--input '{\"prompt\":\"...\"}' --no-wait | jq -r .request_id","category":"blocker","line_end":116,"severity":"low","line_start":115},{"id":"blocker:SKILL.md:151:system-reconnaissance","file":"SKILL.md","pattern":"System reconnaissance","snippet":"| **Rapid concepting → polish workflow** | 4B for exploration, 9B for the final pass — same prompt g","category":"blocker","line_end":151,"severity":"low","line_start":151}],"finding_verdicts":[{"id":"external_commands:SKILL.md:13:ruby-shell-backtick-execution","reason":"The backticks format a RunComfy command in the Markdown description. This is the declared image-generation operation, not executable Ruby or shell substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:14:ruby-shell-backtick-execution","reason":"The backticks delimit the model variant fragment /4b/ in prose. No executable expression or shell substitution appears.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:27:ruby-shell-backtick-execution","reason":"The matched text is a Markdown bash fence around a skill installation example. The command installs the named skill and contains no backtick substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:29:ruby-shell-backtick-execution","reason":"Line 29 closes a Markdown code fence. The match extends across document prose rather than identifying executable backticks.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:49:ruby-shell-backtick-execution","reason":"Inline backticks mark the documented installation of the required RunComfy CLI. No obfuscated command, injected argument, or unrelated execution is shown.","verdict":"false_positive","confidence":0.98},{"id":"external_commands:SKILL.md:50:ruby-shell-backtick-execution","reason":"The documented login command initiates the service authentication flow required for image generation. Its backticks are Markdown formatting, not shell command substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:51:ruby-shell-backtick-execution","reason":"This prerequisite documents a token environment variable as an alternative to interactive login. It contains no actual credential or executable backtick expression.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:57:ruby-shell-backtick-execution","reason":"The heading formats the 9B model identifier as inline code. It does not execute a command.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:63:ruby-shell-backtick-execution","reason":"Backticks identify the prompt field in an input schema table. This is parameter documentation, not executable code.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:64:ruby-shell-backtick-execution","reason":"Backticks format the steps parameter in the schema table. The surrounding text describes generation settings without executing anything.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:65:ruby-shell-backtick-execution","reason":"The width field and image aspect limits are Markdown table content. There is no shell or Ruby execution.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:66:ruby-shell-backtick-execution","reason":"The height and width names are inline-code parameter labels. They are not commands or shell substitutions.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:68:ruby-shell-backtick-execution","reason":"The heading contains the 4B model identifier within Markdown backticks. It is a documented endpoint name, not an executable expression.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:72:ruby-shell-backtick-execution","reason":"The steps field is formatted as inline code in a description of model defaults. No execution occurs.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:76:ruby-shell-backtick-execution","reason":"The inline /edit fragment describes endpoint selection for reference images. It is not a command, and the later backticks open another Markdown fence.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:82:ruby-shell-backtick-execution","reason":"The match is a Markdown fence around an intended image-generation command, not backtick execution. Prompt argument quoting is assessed separately as a semantic concern.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:86:ruby-shell-backtick-execution","reason":"This line closes the concepting example and precedes the next fenced example. Markdown delimiters do not execute shell substitutions.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:90:ruby-shell-backtick-execution","reason":"The backticks open a Markdown example for the declared 9B generation operation. The separate quoting concern does not make this a backtick execution pattern.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:99:ruby-shell-backtick-execution","reason":"The matched backticks close the 9B example and span prose before another fence. They are Markdown syntax, not executable code.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:103:ruby-shell-backtick-execution","reason":"The fence introduces a normal wide-format generation example. No Ruby or shell backtick substitution exists; prompt escaping is addressed separately.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:107:ruby-shell-backtick-execution","reason":"The matched token closes a Markdown code block. Subsequent text documents download behavior rather than executing a backtick expression.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:109:ruby-shell-backtick-execution","reason":"Inline backticks format output host patterns and a CLI option. The prose describes expected artifact downloads, not shell command substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:113:ruby-shell-backtick-execution","reason":"The fenced example submits a generation request and extracts its request identifier with jq. It contains no executable backtick substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:116:ruby-shell-backtick-execution","reason":"This is the closing Markdown fence for the request submission example. The following section is prompting guidance, not executable shell content.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:122:ruby-shell-backtick-execution","reason":"The matched span contains descriptive image prompts and model guidance. Markdown and quoted prompt text do not constitute shell execution.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:135:ruby-shell-backtick-execution","reason":"This span discusses typography, model limitations, and sample prompts. Its formatting does not define executable backtick expressions.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:158:ruby-shell-backtick-execution","reason":"The fence encloses a natural-language image prompt. There is no shell command or Ruby expression inside it.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:163:ruby-shell-backtick-execution","reason":"The closing prompt fence and following product-photo heading are Markdown document structure. They perform no command execution.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:167:ruby-shell-backtick-execution","reason":"The fenced content is a product photography prompt describing a mug and lighting. It is not shell code.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:171:ruby-shell-backtick-execution","reason":"This match crosses a closing prompt fence and the next example heading. Markdown fences do not create executable shell substitutions.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:175:ruby-shell-backtick-execution","reason":"The code fence contains natural-language reference-image editing instructions. No executable command is present.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:179:ruby-shell-backtick-execution","reason":"The fence closes a sample image prompt before the limitations section. This is Markdown formatting, not code execution.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:191:ruby-shell-backtick-execution","reason":"Backticks format the CLI name in an exit-code reference. The table describes operational errors and executes no commands.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:197:ruby-shell-backtick-execution","reason":"The width example is inline schema documentation in an error table. Later text describes invocation; neither contains shell backtick execution.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:206:ruby-shell-backtick-execution","reason":"The inline command documents the skill's declared model invocation. Its Markdown backticks are not evaluated as shell substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:207:ruby-shell-backtick-execution","reason":"Backticks format the documented HTTPS model API endpoint. Authentication to that service is expected behavior, not executable backtick code.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:208:ruby-shell-backtick-execution","reason":"The request_id name and abbreviated GET route are inline documentation of polling. They are not executable shell expressions.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:209:ruby-shell-backtick-execution","reason":"Backticks identify the result route, download host suffixes, and output option. These describe expected artifact retrieval without shell substitution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:210:ruby-shell-backtick-execution","reason":"Ctrl-C and the cancellation route are inline documentation of request cancellation. No executable backtick expression appears.","verdict":"false_positive","confidence":1},{"id":"external_commands:SKILL.md:219:ruby-shell-backtick-execution","reason":"Inline code formats the login command, credential path, and token variable. This documents authentication storage rather than shell backtick execution.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:220:ruby-shell-backtick-execution","reason":"The input option is formatted with Markdown backticks; no backtick execution is present. The unsupported shell-safety assurance is assessed separately.","verdict":"false_positive","confidence":0.99},{"id":"external_commands:SKILL.md:222:ruby-shell-backtick-execution","reason":"The inline values are documented service hostnames and download patterns. They do not execute commands; their implementation was not included for verification.","verdict":"false_positive","confidence":0.99},{"id":"network:SKILL.md:17:hardcoded-url","reason":"The URL is homepage metadata for the named image-generation provider. It does not transmit data or define an executable request.","verdict":"false_positive","confidence":1},{"id":"network:SKILL.md:23:hardcoded-url","reason":"These are provider, model, and source-repository links with referral parameters. They are reference links, not instructions to exfiltrate local data.","verdict":"false_positive","confidence":0.99},{"id":"network:SKILL.md:76:hardcoded-url","reason":"The link points to the model API documentation for reference-image fields. It is a relevant reference, not an unrelated data destination.","verdict":"false_positive","confidence":0.99},{"id":"network:SKILL.md:202:hardcoded-url","reason":"The hardcoded URL is the provider's CLI troubleshooting reference. No upload, credential disclosure, or executable network request is specified.","verdict":"false_positive","confidence":1},{"id":"network:SKILL.md:207:hardcoded-url","reason":"The endpoint is the declared RunComfy generation API, using the user's service token for authentication. This is disclosed task-related communication, not credential exfiltration.","verdict":"false_positive","confidence":0.98},{"id":"filesystem:SKILL.md:208:path-traversal-sequence","reason":"The ellipsis abbreviates the API prefix in GET .../requests/<id>/status. It is not a parent-directory sequence or filesystem access.","verdict":"false_positive","confidence":1},{"id":"filesystem:SKILL.md:209:path-traversal-sequence","reason":"The ellipsis abbreviates an API result route. Downloads are described separately; no relative path traversal or attacker-controlled filename is shown.","verdict":"false_positive","confidence":1},{"id":"filesystem:SKILL.md:210:path-traversal-sequence","reason":"POST .../requests/<id>/cancel is an abbreviated cancellation endpoint. The dots do not navigate local directories.","verdict":"false_positive","confidence":1},{"id":"filesystem:SKILL.md:219:hidden-file-in-home-directory","reason":"The path is the named CLI's documented credential store with owner-only mode 0600. No unrelated credential collection or unauthorized read is instructed.","verdict":"false_positive","confidence":0.98},{"id":"filesystem:SKILL.md:219:hidden-file-access","reason":"The hidden configuration path is disclosed authentication storage for runcomfy login. Its presence alone does not indicate malicious hidden-file access.","verdict":"false_positive","confidence":0.98},{"id":"blocker:SKILL.md:65:at-command-scheduling","reason":"The word at occurs in the phrase describing an aspect ratio capped at 16:9. There is no at scheduler invocation.","verdict":"false_positive","confidence":1},{"id":"blocker:SKILL.md:33:system-reconnaissance","reason":"This paragraph describes iteration latency and product visualization. It contains no command or instruction to inspect the host system.","verdict":"false_positive","confidence":1},{"id":"blocker:SKILL.md:45:system-reconnaissance","reason":"The instruction chooses a generation model from the user's stated preference. It does not collect system information or enumerate the environment.","verdict":"false_positive","confidence":1},{"id":"blocker:SKILL.md:115:system-reconnaissance","reason":"jq extracts request_id from the generation API response. That identifier tracks a submitted job and is not host reconnaissance.","verdict":"false_positive","confidence":0.99},{"id":"blocker:SKILL.md:151:system-reconnaissance","reason":"The table recommends 4B for exploration and 9B for final refinement. It contains no system discovery or information collection.","verdict":"false_positive","confidence":1}],"semantic_findings":[{"title":"Shell Quoting Safety Is Incorrectly Assumed","severity":"medium","locations":[{"file":"SKILL.md","line_end":98,"line_start":82},{"file":"SKILL.md","line_end":106,"line_start":103},{"file":"SKILL.md","line_end":220,"line_start":220}],"confidence":0.88,"description":"Examples place <user prompt> inside single-quoted shell JSON without escaping guidance. Literal substitution of an apostrophe can terminate the shell argument before CLI processing. The statement \"No shell injection surface from prompt content\" overlooks this boundary and can encourage unsafe command construction.","confidence_reasoning":"The examples and absolute safety claim are explicit. Exploitation depends on literal prompt substitution; no executable interpolation implementation was supplied."}],"subject_marketplace_commit_sha":"e4fe64cde5e2867f53e31488fa673df18e2e8130","subject_content_hash":"f808637da840042e899b5186ba35d5aa24bfa5585151e48e822ca67c030b350a","subject_tree_hash":"df7bd662cf42551a752f518066f9856b38c7bdd3c2b06204b3a7ead609c59b4b","subject_plugin_path":"skills/prime-skills/flux-2-klein","audit_payload_hash":"ecdea592cecce1867fab1223e1c87be4","confirmed_risk_level":"medium","scanner_version":"3.0.0","policy_version":"skillstore-security-audit-policy-v1","subject":{"marketplaceCommitSha":"e4fe64cde5e2867f53e31488fa673df18e2e8130","contentHash":"f808637da840042e899b5186ba35d5aa24bfa5585151e48e822ca67c030b350a","treeHash":"df7bd662cf42551a752f518066f9856b38c7bdd3c2b06204b3a7ead609c59b4b","pluginPath":"skills/prime-skills/flux-2-klein","auditPayloadHash":"ecdea592cecce1867fab1223e1c87be4"},"scannerVersion":"3.0.0","policyVersion":"skillstore-security-audit-policy-v1"},"auditTranslation":null,"localization":{"requestedLocale":"en","contentLocale":"en","availableLocales":["en"],"fallbackToEnglish":false},"attestation":{"availability":"issued","url":"/api/skills/prime-skills-flux-2-klein/audits/1/attestation","status":"active"},"trust":{"publicState":"public","auditState":"complete","auditCurrentness":null,"confirmedRiskLevel":"medium","confirmedFindingCount":1,"capabilityReviewCount":0,"needsReviewCount":0,"falsePositiveCount":0,"agentAutoInstallPolicy":"allowed","manualInstallPolicy":"allowed","artifactSignatureState":"available","attestationState":"active","verificationState":"not_verified"},"isLatest":true}}