CtrlK
BlogDocsLog inGet started
Tessl Logo

cicd-secrets-exfil

Extracting CI secrets / OIDC tokens once you have code execution in a build job — echo/printenv exfil, log-masking bypass (base64, char-split, reversal), OIDC token abuse to assume cloud roles, GITHUB_TOKEN / CI_JOB_TOKEN scope abuse, cache / artifact secret leakage, provenance pivot.

65

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/standard/exploit/cicd/cicd-secrets-exfil/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

82%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.

A dense, highly actionable offensive-security skill with copy-paste bash and a solid authorization decision gate. Conciseness and workflow clarity are strong; the only meaningful gap is that validation checkpoints live in a separate advisory section rather than inline with the destructive code paths.

DimensionReasoningScore

Conciseness

Largely lean — code-forward with minimal preamble and no padding about what CI/CD or OIDC 'is'; a few prose explanations (trust-policy enumeration, provenance narrative) could be trimmed but every block is functional.

4 / 5

Actionability

Copy-paste-ready bash across every section (env grep, 8 masking-bypass variants, OIDC mint+assume-role, GITHUB_TOKEN/GitLab job-token API calls, cache/artifact enumeration, cosign signing); commands are concrete and parameterized with the real env vars.

5 / 5

Workflow Clarity

Clear sectioned progression (surface map -> masking -> env -> OIDC -> token abuse -> cache -> provenance -> detection -> decision gate) and a numbered Decision Gate with explicit validation gates (print proof not secret, don't exchange OIDC, dry-run pushes, scrub artifacts); docked one point because the destructive/batch operations lack an inline validate-before-proceed checkpoint inside the operational sections themselves (gates are advisory, separate from the code blocks).

4 / 5

Progressive Disclosure

Well-organized into clearly headed sections with a Tools table and References list; no bundle files exist (references/scripts/assets absent), so there is nothing to split out, and the single-file structure is appropriate for the scope — minor gap only in that References are bare text rather than linkable paths.

4 / 5

Total

17

/

20

Passed

Description

75%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.

Highly specific and distinctive description of CI/CD secret-exfiltration capabilities, written in correct third-person voice. Its main weakness is the absence of an explicit 'Use when...' trigger clause in the description itself, capping completeness.

Suggestions

Add an explicit 'Use when...' clause to the description (e.g., 'Use when you have code execution in a CI build job and need to extract or prove access to CI secrets, OIDC tokens, or CI-scoped tokens').

Surface a few more user-natural trigger phrases in the description rather than relying solely on metadata.when_to_use.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'echo/printenv exfil', 'log-masking bypass (base64, char-split, reversal)', 'OIDC token abuse to assume cloud roles', 'GITHUB_TOKEN / CI_JOB_TOKEN scope abuse', 'cache / artifact secret leakage, provenance pivot' — comprehensive coverage of the domain.

5 / 5

Completeness

The 'what' is clear and concrete, but there is no explicit 'Use when...' clause in the description; trigger guidance lives only in metadata.when_to_use, and per the rubric a missing 'Use when...' clause caps completeness at 3.

3 / 5

Trigger Term Quality

The when_to_use metadata surfaces natural terms ('ci secrets exfil', 'printenv', 'masking bypass', 'oidc', 'id-token', 'aws assume-role', 'github_token', 'ci_job_token', 'cache artifact'), but the description itself leans on jargon over the casual phrases a user might actually say; a few natural synonyms are missing.

4 / 5

Distinctiveness Conflict Risk

Clear narrow niche (CI/CD secret exfiltration and OIDC cloud-role abuse) with distinct triggers unlikely to fire for unrelated skills.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository
PurpleAILAB/Decepticon
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.