CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

bounty-hunting-methodology

Bug bounty white-box hunting methodology. Load when the target is an open-source project with a security advisory program, bug bounty, or responsible disclosure policy.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/standard/analyst/bounty-hunting/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%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 content is a strong, lean instruction skill: a clearly sequenced white-box loop with an explicit validation gate, concrete tooling, and no filler. Its main gaps are the missing validate_finding call signature and a slightly long single-file layout that could offload reference material. The report title convention with good/bad examples is a particularly effective pattern.

DimensionReasoningScore

Conciseness

The body is dense and skimmable: bullet lists of sinks and trust-boundary sources, a one-line find command, and no padding or explanation of concepts Claude already knows. A few interpretive sentences (e.g. "Projects with zero advisories are either very secure or don't have a disclosure process") could be tightened, which keeps it just below the leanest anchor.

4 / 5

Actionability

Guidance is mostly executable: a concrete find command, kg_add_node snippets with props, named tools (semgrep, grep), and explicit validate_finding requirements (poc_command, success_patterns, negative_command, negative_patterns, cvss_vector). Gaps remain — validate_finding is never shown as an actual call signature, and sink discovery says "use semgrep" without a concrete query example.

4 / 5

Workflow Clarity

The five-step core loop (map project, map trust boundaries, crown jewels, trace source to sink, prove or discard) is clearly sequenced, and Step 5 is an explicit validation checkpoint with a working PoC, a negative control run, matching patterns, and a prove-or-discard decision loop ("An unreproduced claim remains unverified"). This matches the anchor for explicit validation steps with feedback loops.

5 / 5

Progressive Disclosure

The body is well-organized into clear sections with no buried or nested references, and the skill ships no bundle files so everything is appropriately self-contained. It exceeds 50 lines and a couple of sections (mechanism families, report title convention) could plausibly be split into reference files, which keeps it below the top anchor for its size.

4 / 5

Total

17

/

20

Passed

Description

65%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 has strong trigger guidance and a clear niche, but the capability statement is a bare domain label. Stating two or three concrete actions (read source, map trust boundaries, validate findings) would raise both specificity and completeness. It is written in appropriate imperative/third-person voice and is concise without padding.

Suggestions

Add 2-3 concrete capabilities after the domain label, e.g. "Maps trust boundaries, traces untrusted input to dangerous sinks, and validates findings with working PoCs."

Include natural trigger synonyms in the description itself, such as "vulnerability research", "security research", or platform names like HackerOne/Bugcrowd, which currently appear only in when_to_use metadata.

State the outcome the skill produces (e.g. triage-ready reports titled by impact and mechanism) so the 'what' answers what the user gets, not just the domain.

DimensionReasoningScore

Specificity

The description names the domain precisely ("Bug bounty white-box hunting methodology") but states no concrete actions — nothing like reading source code, tracing untrusted input, or validating findings. It matches the anchor for naming the domain with minimal/generic actions rather than listing 1-2 concrete capabilities.

2 / 5

Completeness

The 'when' is explicit and specific ("Load when the target is an open-source project with a security advisory program, bug bounty, or responsible disclosure policy"), but the 'what' is only a domain label with no stated capabilities. Both halves are present yet one is underdeveloped, matching the anchor for both present with room to be stronger.

4 / 5

Trigger Term Quality

Natural phrases like "bug bounty", "responsible disclosure", and "security advisory program" are terms users would actually say. Common variations and platform names (HackerOne, Bugcrowd, vulnerability research, security research) are absent from the description itself, so it falls just short of comprehensive coverage.

4 / 5

Distinctiveness Conflict Risk

"Bug bounty white-box hunting methodology" carves a clear niche with distinct triggers tied to advisory/bounty programs. There is minor overlap risk with general security-review and pentest skills, which keeps it below the minimal-conflict anchor.

4 / 5

Total

14

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

PurpleAILAB/Decepticon

Path

packages/decepticon/decepticon/skills/standard/analyst/bounty-hunting/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.