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 commands/dzan314/yapper/yapper-commitgit clone --depth 1 https://github.com/dzan314/yapperWhat 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 | $0.00000 | $0.00781 |
| Opus 5 | $0.00000 | $0.00391 |
| Sonnet 5 | $0.00000 | $0.00156 |
| Haiku 4.5 | $0.00000 | $0.00078 |
Grade A, and why
yapper-commit 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 yesterday.
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 — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/yapper-commit Command
Description
The /yapper-commit command generates a verbose, fully-explained Git commit message
for the current change or a described change. The resulting commit message explains
not just what changed, but why it changed, what problem it solves, and how
the solution works — providing maximum context for future readers of the git history.
Usage
/yapper-commit Generate a verbose commit for staged changes
/yapper-commit [description] Generate a verbose commit for described change
/yapper-commit --diff [paste diff] Generate a verbose commit from a pasted diff
Output Format
The generated commit message MUST follow this structure:
<type>(<scope>): <concise subject line — max 72 characters>
<Body paragraph 1: What changed and at a high level, why.>
<Body paragraph 2: Technical explanation of how the change works.>
<Body paragraph 3: Why this approach was chosen over alternatives, if applicable.>
<Body paragraph 4: Any notable side effects, caveats, or follow-up work required.>
Resolves: #<issue number if known>
Tested: <brief description of how this was tested>
Rules
-
The subject line must be concise (under 72 characters) and use the conventional commits format:
type(scope): description -
The body must contain at least two full paragraphs of prose explanation
-
The what and the why must both be addressed explicitly
-
If the change fixes a bug, the root cause of the bug must be described
-
If the change adds a feature, the use case must be described
-
Technical terms used in the body should be briefly explained if they are not universally understood
-
The tone should be professional but direct — future engineers reading this commit deserve clarity, not poetry (unlike Victorian mode)
Example
Input: "Fixed a bug where users were getting logged out randomly"
Output:
fix(auth): resolve intermittent session expiry caused by stale token cache
This commit addresses a race condition in the session management layer that was
causing authenticated users to be unexpectedly logged out, particularly under
conditions of concurrent API requests. The root cause was that the token validity
check was reading from a cached expiry timestamp rather than evaluating the live
token, meaning that if the token was silently refreshed by a parallel request,
the cached timestamp would still reflect the old (now-invalid) expiry value and
incorrectly trigger the logout flow.
The fix replaces the cached timestamp lookup with a direct evaluation of the
decoded JWT (JSON Web Token — a self-contained, base64-encoded credential that
includes its own expiry claim) on every auth check. This ensures that the expiry
evaluation always reflects the most current token state, regardless of whether
a refresh has occurred on a parallel execution path.
This approach was preferred over a mutex-based lock on the token cache because it
eliminates the cache dependency entirely rather than merely serialising access to
a fundamentally stale data source. The performance impact is negligible: JWT
decoding is a synchronous O(1) operation with no I/O.
Developers integrating with this auth module should note that the token cache has
been removed entirely in this change. Any code reading directly from the cache
(outside of the standard auth middleware) will need to be updated to use the new
`getActiveToken()` utility function introduced in this commit.
Resolves: #412
Tested: Manual session testing across Chrome, Firefox, Safari; concurrent request
simulation using the auth stress test suite in /tests/auth/concurrent.test.ts
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.
- yesterday First seen · 92 lines · 0 tokens per session scan A db725a400202
yapper-commit is a command published in the GitHub repository dzan314/yapper (1 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 781 tokens. 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 commands, from other repositories
hunt
Scan job platforms, score vacancies, generate cover letters, and apply.
council
Run an LLM council — five agents tackle the same task with different roles, then synthesize.
debug
Parallel debugging — five agents propose root-cause hypotheses for a symptom.
review
Parallel code review from all five agents on the current git diff.
second-opinion
Quick independent second opinion from one agent on a decision or approach.
vote
Parallel vote — five agents cast YES / NO / ABSTAIN on a proposition, tally is returned.