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 cyfung1031/skills --skill convergence_gategit clone --depth 1 https://github.com/cyfung1031/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/cyfung1031/skills/convergence_gate)<a href="https://agentmods.dev/skills/cyfung1031/skills/convergence_gate"><img src="https://agentmods.dev/badge/skills/cyfung1031/skills/convergence_gate/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/cyfung1031/skills/convergence_gate"><img src="https://agentmods.dev/badge/skills/cyfung1031/skills/convergence_gate.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00069 | $0.02950 |
| Opus 5 | $0.00034 | $0.01475 |
| Sonnet 5 | $0.00014 | $0.00590 |
| Haiku 4.5 | $0.00007 | $0.00295 |
Grade A, and why
convergence_gate 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 9d 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 — 403 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ConvergenceGate
Purpose
Software can always be improved. Therefore this loop has no natural end:
while possible_improvement_exists:
mutate()
ConvergenceGate replaces it with:
while authorized_completion_blocker_exists:
make_smallest_sufficient_change()
verify_blocker_closed()
stop_mutating()
The governing question is:
What concrete evidence requires another change for the agreed outcome?
This is a termination and scope-control skill. It does not replace code review, testing, debugging, architecture analysis, or human judgment.
Minimal control loop
Keep this loop explicit because it defines termination:
1. DEFINE DONE
What observable conditions make the requested outcome complete?
2. ESTABLISH CURRENT STATE
Is the relevant baseline/spec/upstream/test/review state current enough for the decision?
3. IDENTIFY A BLOCKER
Which completion condition is demonstrably unmet now?
4. AUTHORIZE
What evidence makes another mutation necessary rather than merely attractive?
5. MUTATE MINIMALLY
Make the smallest change that closes that blocker.
6. VERIFY
Run the closure test and required broader verification at the appropriate boundary.
7. CONVERGE
If no authorized blocker remains, stop or hand off for required review.
Do not expand this into a mandatory full audit sequence. Steps with no decision consequence may remain implicit.
Completion contract
Convert the target into finite, observable conditions that can change the final verdict. For coding work, this often includes:
target: <requested outcome>
non_goals: [<explicit exclusions>]
authority: [<user/spec/issue/reviewer/source of truth>]
completion_conditions:
- <observable required behavior>
- <required integration state>
- <required tests/CI>
- <required review/handoff state>
Do not use impossible conditions such as "perfect architecture," "all imaginable edge cases," or "no reviewer could ever find another issue." Adjectives such as "exhaustive," "robust," "production-ready," or "thorough" are not completion conditions by themselves. Translate them into a finite set of observable checks, risk classes, environments, or an explicit review budget before they can authorize continued work.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 403 lines · 69 tokens per session scan A 9e459e22b7d5
convergence_gate is a skill published in the GitHub repository cyfung1031/skills (2 stars, last pushed 29d ago), licensed MIT. It adds 69 tokens to every session and 2,950 once invoked, about $0.0003 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
cw-gates
Use before claiming any Codewhale change is done, green, or ready to land: the focused-to-broad verification ladder, the budget checks CI enforces, and the rules for what counts as a passing test.
writing
A writing guide for turning verified facts and calculations into finished text for a specific audience. It follows the requested language, structure, and length.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
interview
Ask one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.
verify
Exercise the real app/API/CLI and collect observable evidence; tests alone do not count as end-to-end verification.
ccc
This skill should be used when code search is needed (whether explicitly requested or as part of completing a task), when indexing the codebase after changes, or when the user asks about ccc, cocoindex-code, or the codebase index. Trigger phrases include 'search the codebase', 'find code related to', 'update the…