Skills coercion-duress
📦

coercion-duress

Content revision r1 High Risk ⚙️ External commands

Review Coercion and Duress Evidence

Historical evidence reviews can miss forced surrender cues, procedural irregularities, and consent issues. This skill guides Claude, Codex, and Claude Code through a structured JSON-only review workflow.

Supports: Claude Codex Code(CC)
⚠️ 38 Poor

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "coercion-duress" from https://skillstore.io/skills/acskamloops-coercion-duress.md and its manifest at https://skillstore.io/api/skills/acskamloops-coercion-duress/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/acskamloops-coercion-duress.md Signed manifest GET /api/skills/acskamloops-coercion-duress/manifest Signed lockfile GET /api/skills/acskamloops-coercion-duress/lockfile

Test it

Using "coercion-duress". A task batch contains correspondence about a surrender meeting and disputed signatures.

Expected outcome:

The review identifies the document title, date, provenance, reliability, OCR status, exact pressure language, and page anchors for each relevant passage.

Using "coercion-duress". A task has garbled OCR text and no usable document context.

Expected outcome:

The result records an OCR failure recommendation and avoids adding unsupported evidence summaries.

Using "coercion-duress". A completed analysis uses words such as likely and suggests.

Expected outcome:

The pre-flight review marks those terms for removal and replaces them with factual wording tied to direct quotes.

Security Audit

High Risk
v9 • 7/6/2026 Open versioned report

Most static backtick findings are Markdown formatting false positives. The skill still includes command-based workflow steps and a prompt-injection risk because it tells agents to obey system_instructions from JSON task files. Remove delegated system authority and add explicit safeguards around script execution and logging.

1
Files scanned
161
Lines analyzed
7
Review items
0
False positives ignored

Confirmed security concerns (1)

High
SKILL.md:20
Prompt Injection Attempt Detected
The skill says to follow the system_instructions block injected into every JSON task file as hard constraints. This lets untrusted task data override agent behavior.
The wording explicitly delegates instruction authority to a block injected into task JSON. That is a direct prompt-injection risk because task data is untrusted.
Capability review items (7)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium
SKILL.md:10
Ruby/shell backtick execution
- For court audit trails, run batches via `codex_exec_runner.sh` with `PUKAIST_CODEX_LOG_EVENTS=1` t
The instruction tells the agent to run codex_exec_runner.sh with event logging. Running repository scripts and writing raw JSONL logs can execute local code and expose task data.
Medium
SKILL.md:35-37
Ruby/shell backtick execution
```powershell
The fenced block contains a command to run refinement_workflow.py get-task. Executing repository workflow scripts from a skill is real external command use.
Medium
SKILL.md:42-44
Ruby/shell backtick execution
```powershell
The fenced block runs a Python one-liner against a task file path. It is intended for local JSON reading, but it still instructs command execution.
Medium
SKILL.md:124-126
Ruby/shell backtick execution
```powershell
The fenced block instructs running submit-task, which appends analysis to an internal report. That can modify repository workflow state.
Medium
SKILL.md:133-135
Ruby/shell backtick execution
```powershell
The fenced block instructs running flag-task for workflow triage. This is local command execution that can update tracking files.
Medium
SKILL.md:137-138
Ruby/shell backtick execution
* **Action:** This command will **automatically move the source file** to the Vision Pipeline (`07
The text says the OCR failure action will automatically move a source file to another pipeline. Automatic file movement is a real workflow mutation risk.
Medium
SKILL.md:138-140
Ruby/shell backtick execution
```powershell
The fenced block instructs running flag-task with OCR_Failure, tied to moving files into the Vision pipeline. That is external command use with filesystem impact.

Risk Factors

⚙️ External commands (41)
SKILL.md:7 SKILL.md:8 SKILL.md:10 SKILL.md:20 SKILL.md:30 SKILL.md:31 SKILL.md:35-37 SKILL.md:37-40 SKILL.md:40-42 SKILL.md:42-44 SKILL.md:44-62 SKILL.md:62 SKILL.md:63-94 SKILL.md:94 SKILL.md:97-99 SKILL.md:99-102 SKILL.md:102-104 SKILL.md:104-109 SKILL.md:109-112 SKILL.md:112 SKILL.md:119 SKILL.md:120 SKILL.md:121-124 SKILL.md:124-126 SKILL.md:126-127 SKILL.md:127-128 SKILL.md:128 SKILL.md:132-133 SKILL.md:133-135 SKILL.md:135-137 SKILL.md:137-138 SKILL.md:138-140 SKILL.md:140-143 SKILL.md:143 SKILL.md:144-145 SKILL.md:145 SKILL.md:35 SKILL.md:42 SKILL.md:124 SKILL.md:133 SKILL.md:138
Audited by: codex View Audit History →
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/acskamloops-coercion-duress/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/acskamloops-coercion-duress/security.svg)](https://skillstore.io/skills/acskamloops-coercion-duress?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/acskamloops-coercion-duress?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/acskamloops-coercion-duress/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/acskamloops-coercion-duress.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA · BibTeX · CFF)

APA citation

ACSKamloops. (2026). coercion-duress security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/acskamloops-coercion-duress/audits/9

BibTeX citation

@techreport{acskamloops-acskamloops-coercion-duress-2026, author = {ACSKamloops}, title = {coercion-duress security audit report (audit version 9)}, institution = {Skillstore}, year = {2026}, number = {9}, url = {https://skillstore.io/skills/acskamloops-coercion-duress/audits/9}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "coercion-duress security audit report (audit version 9)" version: "unspecified" type: report authors: - name: "ACSKamloops" date-released: "2026-07-06" url: "https://skillstore.io/skills/acskamloops-coercion-duress/audits/9" identifiers: - type: other value: "skillstore:acskamloops-coercion-duress:audit:9" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium
55
Architecture
85
Maintainability
87
Content
70
Community
83
Spec Compliance

What You Can Build

Index forced surrender evidence

Review task batches for threats, pressure language, and consent problems while preserving exact quotes.

Prepare neutral evidence summaries

Convert document excerpts into factual summaries with provenance, reliability, OCR status, and page references.

Validate Coercion_Duress submissions

Check metadata, banned words, date format, and manager review status before workflow submission.

Try These Prompts

Review one task batch 
Use the coercion-duress skill to review one Coercion_Duress task batch. Extract only factual evidence about threats, pressure, consent, and procedure.
Build citation-ready summaries 
Review this batch and prepare neutral summaries with document identifiers, dates, provenance, reliability, OCR status, verbatim quotes, and page references.
Flag workflow exceptions 
Review the task set for corrupt text, irrelevant material, OCR failure, or provenance failure. State the proper flag reason for each exception.
Audit before submission 
Audit the completed Coercion_Duress analysis against the skill rules. Find missing metadata, opinion words, weak citations, short summaries, and premature conclusions.

Best Practices

  • Read the full super task before judging relevance or extracting evidence.
  • Use verbatim quotes and page anchors when the task includes usable evidence.
  • Run pre-flight checks for metadata, date format, length, and opinion words before submission.

Avoid

  • Using keyword matches instead of reading the surrounding context.
  • Returning Unknown for metadata that appears in the task text.
  • Treating workflow submissions as final before manager approval.

Frequently Asked Questions

What documents does this skill review?
It reviews Coercion_Duress JSON task batches that contain historical document excerpts and metadata.
Does it provide legal conclusions?
No. It requires neutral summaries, verbatim evidence, and factual context only.
Can it analyze PDFs directly?
No. The instructions limit the workflow to JSON input and JSON output.
What evidence does it prioritize?
It prioritizes threats, forced surrender language, consent problems, translation issues, and procedural irregularities.
Why are metadata checks required?
Metadata checks keep document identifiers, dates, provenance, reliability, and OCR status consistent for later review.
Does it need local scripts?
Yes. The described fetch, submit, and flag workflow depends on repository scripts.

Developer Details

License

MIT

Skillstore revision

r1

Version notice

The author did not declare a version.

Repository

https://github.com/ACSKamloops/shs-engine/tree/master/.codex/skills/coercion-duress

Ref

c1fdca50ff516318f65fed0d7f9e82797c5171dc

Maintenance freshness

7/18/2026

Usage

7 downloads · 181 views

File structure

📄 SKILL.md