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 agentmods add skills/kucherenko/gangsta/resource-developmentnpx skills add kucherenko/gangsta --skill resource-developmentgit 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/resource-development)<a href="https://agentmods.dev/skills/kucherenko/gangsta/resource-development"><img src="https://agentmods.dev/badge/skills/kucherenko/gangsta/resource-development.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00042 | $0.00986 |
| Opus 5 | $0.00021 | $0.00493 |
| Sonnet 5 | $0.00008 | $0.00197 |
| Haiku 4.5 | $0.00004 | $0.00099 |
Grade A, and why
resource-development 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 6d 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Resource Development: Infrastructure and Tooling
Overview
The Underboss decomposes the signed Contract into Work Packages, sets up infrastructure for parallel execution, and produces the Execution Plan — the detailed task breakdown that governs The Hit.
Trigger
Invoked after the Don signs the Contract (The Sit-Down complete).
Process
Step 1: Decompose Contract into Work Packages
For each requirement in the Contract:
- Identify affected files (create/modify)
- Determine territory ownership (which Crew Lead)
- Define acceptance criteria (from Contract)
- Write verification command
- Estimate token budget
Each Work Package must be:
- Independent — Can be implemented without waiting for other Work Packages (within the same territory, order may matter)
- Verifiable — Has a specific test or command that proves it works
- Bite-sized — 2-5 minutes of Worker tasks
- Traceable — References a specific Contract clause
Step 2: Define Territories
Group Work Packages by domain and define Crew Lead territories:
## Territory: <Name>
**Domain:** <Description>
**Files:** <Glob patterns>
**Work Packages:** WP-001, WP-003, WP-007
**Workers:** <Parallel count>
**Budget:** <Token allocation>
Step 3: Set Up Isolation
If the project uses git:
- Create a branch for the Heist:
heist/<heist-name> - Optionally create worktrees per territory for true parallel work
If not:
- Document the isolation strategy (directory copies, etc.)
Step 4: Verify Prerequisites
Before The Hit begins:
- All existing tests pass on the current baseline
- Required dependencies are available
- File paths in Work Packages are valid
- No merge conflicts with the base branch
Step 5: Produce Execution Plan
Compile everything into the Execution Plan.
Execution Plan Format
Save to: docs/gangsta/<heist-name>/plans/YYYY-MM-DD-execution-plan.md
---
heist: <heist-name>
date: YYYY-MM-DD
total-work-packages: <N>
territories: <N>
estimated-total-budget: <tokens>
---
# Execution Plan: <Heist Name>
## Territories
### Territory: <Name 1>
**Crew Lead Domain:** <description>
**Files:** <globs>
**Workers:** <N parallel>
**Budget:** <tokens>
### Territory: <Name 2>
...
## Work Packages
**Prevention Guidance (all Work Packages):** Do NOT reproduce Gangsta-internal spec identifiers (FR-xxx, NFR-xxx, WP-xxx, where xxx is one or more digits) in any deliverable — source code, test files, code comments, or documentation. These identifiers belong only in `docs/gangsta/`.
### WP-001: <Title>
**Territory:** <Name>
**Contract Clause:** <FR-NNN>
**Files:**
- Create: `path/to/file`
- Test: `tests/path/to/test`
**Acceptance Criteria:**
1. <criterion>
**Verification:** `<command>`
**Budget:** <tokens>
**Dependencies:** None | WP-XXX
### WP-002: <Title>
...
## Execution Order
1. Independent packages (can run in parallel): WP-001, WP-002, WP-005
2. Depends on group 1: WP-003 (needs WP-001)
3. ...
## Baseline Verification
- Tests: <PASS/FAIL — must be PASS>
- Dependencies: <OK/MISSING>
- Branch: heist/<heist-name> created from <base>
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.
- 6d ago First seen · 146 lines · 42 tokens per session scan A a3cea4fd6c4c
resource-development is a skill published in the GitHub repository kucherenko/gangsta (81 stars, last pushed 22d ago), licensed MIT. It adds 42 tokens to every session and 986 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
cocotrace
Build and query the SHA-256 requirements-to-implementation traceability graph across lifecycle artifacts.
lean-review
CocoLean diff-scoped over-engineering audit — scans uncommitted git diff and applies five classification tags (delete/stdlib/native/yagni/shrink) to identify unnecessary surface area before commit.
cocoplus-config
CocoPlus configuration SSOT — $cocoplus sync propagates cocoplus.toml into downstream artifacts; $cocoplus migrate-config converts legacy safety-config.json. Invoked via $cocoplus sync and $cocoplus migrate-config.
bloom-crystallize
CocoBloom crystallization — converts a successful ship execution trace into a reusable skill file. Requires recent successful $ship. Invoked via $bloom crystallize.
context-add
Guided wizard to capture or update organizational Snowflake/Cortex standards into .cocoplus/context/ .md. Presents a menu of 6 standard types, collects answers via multi-turn dialogue, and commits the file.
discuss-red-team
CocoDiscuss Red-Team Mode — adversarial devil's advocate session that challenges spec assumptions before $plan. Invoked via $discuss --red-team.