CtrlK
BlogDocsLog inGet started
Tessl Logo

competition-prompt-injection

Internal downstream skill for ctf-sandbox-orchestrator. CTF-sandbox workflow for prompt-injection, retrieval poisoning, memory contamination, planner drift, MCP or tool-boundary abuse, and agent exfiltration challenges. Use when the user asks to analyze prompt injection, retrieval poisoning, memory contamination, planner drift, tool-argument corruption, or secret exposure caused by an agent chain. Use only after `$ctf-sandbox-orchestrator` has already established sandbox assumptions and routed here.

71

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tight, well-structured instruction-only methodology that assumes competence and offloads the checklist to a clearly signaled one-level-deep reference. It falls just short of top marks on actionability and workflow clarity because it lacks a worked minimal-exploit example and an explicit error-recovery loop.

Suggestions

Add one compact worked example showing a minimal exploit chain (untrusted chunk -> rewritten planner state -> final tool invocation) so the actionability guidance is fully concrete.

Insert an explicit feedback loop in the Workflow, e.g. 'If the chain does not reproduce a side effect, narrow the transcript and re-prove before exploring variants', to reach the validation/checkpoint anchor.

Link the reference with markdown syntax (references/prompt-injection.md) in addition to the backtick form to make navigation unambiguous.

DimensionReasoningScore

Conciseness

The body is lean bullet-and-short-prose guidance with no concept padding (it never explains what prompt injection is) and assumes Claude's domain competence, so every line earns its place per the lean-and-efficient anchor.

5 / 5

Actionability

Directives like 'Identify the first untrusted content that becomes model-visible', 'Record the exact point where text becomes a tool argument, file path, network target, or secret request', and 'Prove one minimal exploit chain before exploring variants' are concrete and actionable, but no worked transcript example illustrates a minimal chain, leaving a minor gap versus the fully-concrete 5 anchor.

4 / 5

Workflow Clarity

A numbered Quick Start (1-5) and a three-phase Workflow give a clear sequence with a proof checkpoint ('Prove one minimal exploit chain before exploring variants') and a side-effect check ('Separate instruction drift from actual side effect'), but there is no explicit error-recovery feedback loop for when a chain fails to reproduce, which is the missing checkpoint versus the 5 anchor.

4 / 5

Progressive Disclosure

The body is a concise overview and points to a single one-level-deep reference via a dedicated 'Read This Reference' section that states its contents ('checklist, evidence layout, and common prompt-boundary pitfalls'); the reference file does not nest further, matching the clear-overview-with-well-signaled-one-level-deep-references anchor.

5 / 5

Total

18

/

20

Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it explicitly covers both what the skill does and when to use it, names a comprehensive set of challenge types, and is gated to a parent skill to minimize conflict. Specificity and trigger-term quality stop just short of the top anchor because the action verbs are generic and a few natural synonyms (e.g. 'jailbreak') are missing.

Suggestions

Add a concrete action verb beyond 'workflow for'/'analyze' (e.g. 'Trace, reproduce, and report exploit chains across ...') to lift specificity toward the comprehensive-action anchor.

Include widely-used user synonyms such as 'jailbreak' or 'indirect prompt injection' alongside the existing trigger terms to broaden natural keyword coverage.

Optionally name the output artifact (e.g. '... and produce a boundary-failure report') so the what clause describes a tangible deliverable.

DimensionReasoningScore

Specificity

Lists many concrete challenge types (prompt-injection, retrieval poisoning, memory contamination, planner drift, MCP/tool-boundary abuse, agent exfiltration, tool-argument corruption, secret exposure) but the action verbs are thin ('workflow for', 'analyze'), leaving minor gaps versus the comprehensive-action anchor at 5.

4 / 5

Completeness

It explicitly states both what ('CTF-sandbox workflow for ... challenges') and when ('Use when the user asks to analyze ...') with concrete trigger phrases, plus an explicit gating condition, matching the clearly-and-explicitly-both anchor.

5 / 5

Trigger Term Quality

The 'Use when the user asks to analyze prompt injection, retrieval poisoning, memory contamination, planner drift, tool-argument corruption, or secret exposure' clause gives good keyword coverage with variations, but common user synonyms such as 'jailbreak' are absent and several terms are specialized jargon, stopping short of the comprehensive 5 anchor.

4 / 5

Distinctiveness Conflict Risk

A narrow agent trust-boundary niche with distinct triggers and an explicit 'Use only after $ctf-sandbox-orchestrator has already established sandbox assumptions and routed here' gate yields minimal conflict risk, matching the clear-niche anchor.

5 / 5

Total

18

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
zhaoxuya520/reverse-skill
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.