Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add kucherenko/gangsta --skill the-hitgit clone --depth 1 https://github.com/kucherenko/gangstaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kucherenko/gangsta/the-hit)<a href="https://agentmods.dev/skills/kucherenko/gangsta/the-hit"><img src="https://agentmods.dev/badge/skills/kucherenko/gangsta/the-hit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kucherenko/gangsta/the-hit"><img src="https://agentmods.dev/badge/skills/kucherenko/gangsta/the-hit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00039 | $0.01776 |
| Opus 5 | $0.00019 | $0.00888 |
| Sonnet 5 | $0.00008 | $0.00355 |
| Haiku 4.5 | $0.00004 | $0.00178 |
Grade A, and why
the-hit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The Hit: Execution and Parallel Coding
Overview
Workers and Crew Leads implement the Work Packages defined in the Execution Plan. Crews work simultaneously on their assigned territories, providing Reports (status updates) to the Underboss.
Trigger
Invoked after the Don approves the Execution Plan (Resource Development complete).
Process
Step 1: Underboss Distributes Work Packages
For each territory, the Underboss sends the Crew Lead:
- The territory definition
- The assigned Work Packages (ordered by dependency)
- The relevant Contract sections
- The applicable Constitution rules
Step 2: Crew Leads Dispatch Workers
Each Crew Lead (invoke gangsta:the-capo) processes their Work Packages:
- For each independent Work Package, dispatch a Worker subagent with
subagent_type: "soldier"(do NOT use"general"or"general-purpose"— these are not valid in a Gangsta Agents installation), providing:- The Work Package brief
- The Contract clause being implemented
- Applicable Negative Constraints
- Territory conventions
- Prevention reminder: Workers MUST NOT reproduce Gangsta-internal spec identifiers (FR-xxx, NFR-xxx, WP-xxx, where xxx is one or more digits) in source code, tests, comments, or documentation. This is prevention guidance, not a gate.
- Technical-debt prevention brief (Laundering gate): Laundering will block the Heist on five technical-debt checks. Workers must build today to pass those checks tomorrow:
- Copy/paste: do not clone build_setup in 5+ lines across files. If two call sites share logic, extract a helper or parameterize the variation. The duplicate will be caught by jscpd (min-lines 5, min-tokens 35) in Laundering.
- Long files: target ≤ 300 source lines per file. Hard block at 500. If a Work Package forces a long file, the Contract must name it and the Don must pre-approve — otherwise split by responsibility.
- Test coverage: write tests for every new public function and every modified branch. Laundering compares against the pre-Heist coverage baseline; coverage on modified code may not drop. If the project has a coverage threshold in its config, the run must meet it.
- Clean architecture: imports flow downward through layers (domain → application → infrastructure → delivery); no business logic in controllers, route handlers, DB adapters, or network clients; one responsibility per file.
- Clean code: remove dead code before reporting; functions ≤ 50 lines (blank/comment excluded); nesting ≤ 4 levels via early returns; no magic numbers in production paths; identifiers must name themselves.
- Worker Report must self-report against these five points. See Step 4.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 154 lines · 39 tokens per session scan A 42ada28d743a
the-hit is a skill published in the GitHub repository kucherenko/gangsta (81 stars, last pushed 27d ago), licensed MIT. It adds 39 tokens to every session and 1,776 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
contract
Outcome-driven Cortex function development — declares a behavioral contract before generation begins, enforces evidence-tiered proof before $ship, and defends against the self-oracle evaluation failure mode.
workflow-patterns
Systematic task implementation using TDD, phase checkpoints, and structured commits. Ensures quality through red-green-refactor cycles, 80% coverage gates, and verification protocols before proceeding.
workflow-patterns
Use this skill when implementing tasks according to Conductor's TDD workflow, handling phase checkpoints, managing git commits for tasks, or understanding the verification protocol.
azure-cosmos-db-py
Build Azure Cosmos DB NoSQL services with Python/FastAPI following production-grade patterns. Use when implementing database client setup with dual auth (DefaultAzureCredential + emulator), service layer classes with CRUD operations, partition key strategies, parameterized queries, or TDD patterns for Cosmos. Triggers…
create-tooluniverse-skill
Create high-quality ToolUniverse skills following test-driven, implementation-agnostic methodology.
review
Enter the Review phase of CocoBrew. Aggregates findings from Code Quality Advisor, CocoCupper intelligence, spec compliance check, and decision coverage gate. Produces review.md with decision points. Requires developer approval before $ship can proceed.