CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

vulnresearch-orchestrator

Five-stage modular vulnerability pipeline orchestrator. Delegates scan → detect → verify → patch → exploit through OPPLAN objectives. Load at orchestrator startup.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

90%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 lean, highly actionable orchestrator skill: concrete delegation templates, explicit batch ceilings, precondition-gated stage sequencing, and built-in validation checkpoints (kg_stats, patch_verify). The only meaningful gaps are the absence of error-recovery guidance for failed stages and no progressive-disclosure structure, though the latter is arguably unnecessary at this size.

DimensionReasoningScore

Conciseness

The ~80-line body is lean and dense with operational rules — dependency gates, batch ceilings, delegation templates, and a report format — with no explanations of concepts Claude already knows and no padding; every token earns its place.

5 / 5

Actionability

The five task() delegation examples are copy-paste-ready with concrete parameters (shard_total=8, kind=candidate, status=pending, validated!=True, patch_verify, /workspace/exploits/<chain_id>/) and cover all pipeline stages, matching "fully executable; specific examples cover the common cases".

5 / 5

Workflow Clarity

The stage dependency diagram gives a clear sequence with explicit precondition gating ("Launch a stage ONLY when its preconditions are met") and validation checkpoints (kg_stats graph-delta checks between stages, patch_verify confirmations), so the batch-operation cap at 3 does not apply. However, no error-recovery feedback loop is specified (what to do when a stage fails or validation rejects a finding), which is required for the 5 anchor.

4 / 5

Progressive Disclosure

Sections are well-organized and nothing that belongs in a separate file is inlined; the ~80-line self-contained body is appropriately placed. It is not a 5 because the skill exceeds the under-50-line simple-skill case and offers no reference structure at all, sitting at "good structure; most content appropriately placed; minor organization gaps".

4 / 5

Total

18

/

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.

A solid description that clearly states what the skill does (five-stage delegation pipeline) with concrete stage names and an explicit load-time trigger. It falls short of top marks because the "when" is a terse system directive rather than user-need trigger phrases, natural synonyms are absent, and internal jargon ("OPPLAN objectives") is unexplained.

Suggestions

Replace "Load at orchestrator startup" with an explicit user-need trigger clause, e.g., "Use when running the vulnerability research pipeline end-to-end or when coordinating scanner/detector/verifier/patcher/exploiter stages" (completeness).

Add natural synonyms users would actually say, such as "vulnerability research", "security findings", or "pentest pipeline", to broaden trigger coverage (trigger_term_quality).

Drop or briefly gloss internal jargon like "OPPLAN objectives" so every stated action is concrete and self-explanatory (specificity).

DimensionReasoningScore

Specificity

The description lists several concrete actions ("Delegates scan → detect → verify → patch → exploit") covering the pipeline stages, but "OPPLAN objectives" is unexplained jargon and no outputs or downstream actions are stated, leaving minor gaps in coverage rather than the comprehensiveness of a 5.

4 / 5

Completeness

The "what" is clear (orchestrates/delegates the five-stage vulnerability pipeline) and a "when" is explicitly present ("Load at orchestrator startup"), so it is not merely implied and avoids the ≤3 cap. However, the when-clause is a terse load directive rather than concrete user-need trigger phrases, fitting the anchor "'when' could be more explicit or specific" rather than the fully explicit 5.

4 / 5

Trigger Term Quality

Good keyword coverage of domain-natural terms ("vulnerability", "pipeline", "orchestrator", "scan", "detect", "verify", "patch", "exploit"), but common synonyms and variations a user might say (e.g., "security research", "pentest", "vulns", "findings") are missing, matching "a few natural terms missing" rather than comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

The description carves a clear niche as a top-level orchestrator that only delegates, but the stage verbs (scan, detect, verify, patch, exploit) are shared with the individual stage skills it names, creating minor overlap risk with closely related skills rather than the minimal conflict risk of a 5.

4 / 5

Total

16

/

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/plugins/vulnresearch/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.