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 vignesh2027/AI-AGENT-SKILLS --skill source-driven-developmentgit clone --depth 1 https://github.com/vignesh2027/AI-AGENT-SKILLSWrote 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/vignesh2027/ai-agent-skills/source-driven-development)<a href="https://agentmods.dev/skills/vignesh2027/ai-agent-skills/source-driven-development"><img src="https://agentmods.dev/badge/skills/vignesh2027/ai-agent-skills/source-driven-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 | $0.00019 | $0.00404 |
| Opus 5 | $0.00010 | $0.00202 |
| Sonnet 5 | $0.00004 | $0.00081 |
| Haiku 4.5 | $0.00002 | $0.00040 |
Grade A, and why
source-driven-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 7d 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.
What it actually says
Overview
Assumptions are expensive. This skill requires every non-obvious implementation decision to be traceable to an authoritative source: the spec, the RFC, the official documentation, the ticket. Without a source, it's a guess.
When to Use
- When implementing a standard or protocol (OAuth, HTTP, JWT, etc.)
- When integrating with a third-party API
- When the implementation relies on undocumented behavior
- When "I think this is how it works" appears in your reasoning
Process
Step 1: Identify what you're implementing
What standard, protocol, API, or requirement are you implementing? What is the authoritative source?
Step 2: Read the source before writing code
For protocols: read the RFC. For APIs: read the official API documentation (not a blog post). For requirements: re-read the spec.
Step 3: Cite the source in the implementation
For non-obvious behavior: add a comment citing the source. Example:
# Per RFC 7519 Section 4.1.4, exp claim is NumericDate (seconds since epoch)
# https://datatracker.ietf.org/doc/html/rfc7519#section-4.1.4
Step 4: Identify the gaps
What does the source not specify? These are places where you must make explicit decisions. Document them as decisions, not silently.
Step 5: Validate against the source
After implementation: validate your implementation against the authoritative source. For protocols: test against a reference implementation.
Verification Requirements
- Authoritative source identified and read before implementation
- Non-obvious behavior cites the source
- Gaps from the source documented as explicit decisions
- Implementation validated against the source
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.
- 7d ago First seen · 46 lines · 19 tokens per session scan A a871f4a6912a
source-driven-development is a skill published in the GitHub repository vignesh2027/AI-AGENT-SKILLS (2 stars, last pushed 9d ago), licensed MIT. It adds 19 tokens to every session and 404 once invoked, about $0.0001 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 2026-08-31.
Other skills, from other repositories
telegram
Owner-only Telegram text bridge and Mini App gateway for the existing Ouroboros interface.
claudexor
Use the local Claudexor control plane for harness-agnostic coding work across Claude Code, Codex, Cursor, and OpenCode. Use when a task benefits from route-aware harness and account selection, quota-aware account rotation, shared thread context, read-only planning or research, best-of-N execution, or cross-harness…
openai-docs
Use when the user asks how to build with OpenAI products or APIs, asks about Codex itself or choosing Codex surfaces, needs up-to-date official documentation with citations, help choosing the latest model for a use case, latest/current/default-model prompting guidance, or model upgrade and prompt-upgrade guidance; use…
knowledge
The owner's knowledge base: a markdown folder that is also a typed graph of people, projects, decisions and terms, via the kb CLI. Use BEFORE answering anything about the owner, who they work with, what a project or term means, or what was decided and why; record durable facts unasked. Not for facts about the code.
Gmail, Google Calendar, Drive, Docs, Sheets and Contacts for a connected Google account: read and search mail, send and reply, book and move meetings, find files, read and write Docs and Sheets, look people up. Use whenever the user mentions their email, inbox, calendar, meetings, Drive, a Google Doc or Sheet.
wallet
Pay x402-payable endpoints in USDC from the sandbox wallet via the wallet CLI, under the owner's spending policy. Use when an API answers 402 Payment Required with an x402 challenge or a task needs a paid service: check the price, then ask; the owner approves each payment outside their auto-approve band.