CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

web-cookie-audit

Cookie-conditional sink discovery — bisect required cookies per sink, session-write timeline for race-condition challenges.

Quality

81%

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

92%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, highly actionable recon procedure: concrete enumeration and bisect steps, exact output table formats with worked examples, and enforced handoff requirements with explicit failure loops. The only trimmable content is a couple of motivational sentences in the intro.

DimensionReasoningScore

Conciseness

The body is efficient: every procedure step is concrete, both tables carry real example rows, and there is no explanation of concepts Claude already knows. It falls just short of anchor 5 because a few sentences are motivational rather than instructional ('Skipping this step is the most common cause of exploit handing back a false "blind sink" verdict') and could be trimmed.

4 / 5

Actionability

For an instruction-only skill the guidance is fully concrete: exact enumeration steps (record every Set-Cookie with attributes and source endpoint, authenticate at each tier), an exact bisect procedure (drop one cookie at a time, re-probe), and copy-paste-ready output templates — the sink-preconditions table, the session-write timeline table with a filled example, and the verbatim 'Required session state' handoff line. This matches anchor 5; it is not 4 because no executable step is left underspecified.

5 / 5

Workflow Clarity

Steps 1–5 are clearly sequenced with embedded validation checkpoints: the bisect loop terminates on an explicit observable ('The cookie whose removal silences the sink is the gating cookie'), the race-condition tag check gates the session-write table, and the handoff is verified by a mandatory 'Required session state' line with a defined failure loop (exploit must re-run cookie enumeration if it is missing). This matches anchor 5's explicit validation steps and feedback loops; the procedure is recon (non-destructive), so the batch-operation cap does not apply.

5 / 5

Progressive Disclosure

The skill is under 50 lines, single-purpose, has no bundle files to reference, and is well organized into an intro plus a numbered Procedure with clearly signaled output tables. Per the scoring note for simple skills, this earns a 5 on well-organized sections alone; the only mentioned files (recon_notes.md, SUMMARY.md) are the recon phase's own output artifacts, not missing bundle references.

5 / 5

Total

19

/

20

Passed

Description

62%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 specific, distinctive, and clearly states what the skill does, but it lacks an explicit 'Use when...' trigger clause and leans on technical jargon rather than natural trigger terms. Moving the trigger phrases already present in metadata.when_to_use into the description would substantially raise completeness and trigger-term quality.

Suggestions

Add an explicit 'Use when...' clause to the description field itself, e.g. 'Use when auditing session cookies, testing sink preconditions, or when a candidate sink appears unreachable without cookies.'

Include natural trigger terms a user would actually say — 'session cookies', 'cookie audit', 'gating cookies', 'sink unreachable' — alongside the jargon terms ('bisect', 'sink discovery').

Mention the auth-tier enumeration and handoff-line outputs in the description so the stated capabilities match what the procedure actually covers.

DimensionReasoningScore

Specificity

The description lists several concrete, domain-specific actions — 'Cookie-conditional sink discovery', 'bisect required cookies per sink', 'session-write timeline' — but omits other capabilities the skill covers (auth-tier cookie-jar enumeration, handoff documentation), leaving minor coverage gaps. This sits between anchor 4 ('several specific actions; minor gaps') and anchor 5 (comprehensive coverage), so 4 is the best fit.

4 / 5

Completeness

The 'what' is clear (discover cookie-conditional sinks, bisect required cookies, build a session-write timeline), but the description field contains no explicit 'Use when...' trigger clause — only a weak implicit 'for race-condition challenges'. Per the judging guideline, a missing 'Use when...' clause caps completeness at 3, and this skill's explicit trigger guidance lives in metadata.when_to_use rather than the description itself.

3 / 5

Trigger Term Quality

Relevant keywords are present ('cookie', 'session', 'race-condition', 'sink'), but the phrasing is jargon-heavy ('bisect', 'cookie-conditional', 'sink discovery') and misses the natural phrases a user would say, such as 'session cookies', 'cookie audit', 'gating cookies', or 'sink unreachable'. Not 2 because the keywords are domain-relevant rather than generic; not 4 because common natural variations and synonyms are absent from the description field itself.

3 / 5

Distinctiveness Conflict Risk

The description carves out a clear niche (cookie-gated sink discovery with per-cookie bisection and session-write timelines) with distinct triggers that no generic web-recon or cookie-handling skill would claim, so conflict risk is minimal. It clearly matches anchor 5 rather than anchor 4, since even closely related skills would not trigger on these phrases.

5 / 5

Total

15

/

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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

PurpleAILAB/Decepticon

Path

packages/decepticon/decepticon/skills/standard/recon/web-recon/cookie-audit/SKILL.md

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.