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 tmj-90/gaffer --skill prepare-digest-deltagit clone --depth 1 https://github.com/tmj-90/gafferWrote 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/tmj-90/gaffer/prepare-digest-delta)<a href="https://agentmods.dev/skills/tmj-90/gaffer/prepare-digest-delta"><img src="https://agentmods.dev/badge/skills/tmj-90/gaffer/prepare-digest-delta/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/tmj-90/gaffer/prepare-digest-delta"><img src="https://agentmods.dev/badge/skills/tmj-90/gaffer/prepare-digest-delta.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 | $0.00116 | $0.00943 |
| Opus 5 | $0.00058 | $0.00472 |
| Sonnet 5 | $0.00023 | $0.00189 |
| Haiku 4.5 | $0.00012 | $0.00094 |
Grade A, and why
prepare-digest-delta 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prepare the Repo Digest delta (apply-at-merge)
You just implemented and evidenced a ticket. You already know exactly what changed — so you are the cheapest place to write down how the Repo Digest should move and which feature this ticket ships. This skill records that as a single, structured evidence row. It is INERT: nothing is applied now. The factory's merge step reads this row post-review and applies it deterministically (no new agent call). If the ticket is rejected, it never merges, so this prepared delta never pollutes the digest.
This is the prepare half of prepare-at-delivery / apply-at-merge. Recording the
delta here — instead of running a fresh claude -p on every merge — is the whole cost
win. Keep it cheap: you are summarising a diff you already have, not re-deriving it.
Steps
-
Decide what the digest should say now. For each Repo Digest section your change makes stale (architecture, key flows, conventions, surface area, gotchas, …), write the SHORT updated prose for that section — only sections you actually changed. If the change is small and touches no section's narrative, you may record zero sections (the merge will still stamp freshness).
-
Name the feature this ticket ships. One feature note: its
name, a one-linesummary, and — if you know it — thescopeNodeandprovenance(e.g. the epic ref). This is what the merge advances/adds toshipped. -
Record ONE evidence row via the Dispatch MCP delivery-evidence path (
attach_delivery_evidence/ therecord-evidenceflow),evidence_type: manual_note, whose summary is exactly the marker line below — the marker, a single space, then a compact one-line JSON payload:GAFFER_DIGEST_DELTA_V1 {"repo":"<repo-name>","sections":[{"section":"<digest section>","content":"<updated prose>"}],"feature":{"name":"<feature name>","summary":"<one line>","scopeNode":"<id or omit>","provenance":"<epic ref or omit>"}}- The summary MUST start with
GAFFER_DIGEST_DELTA_V1(marker + one space). - The payload MUST be valid one-line JSON.
sectionsmay be[];featuremay be omitted if this ticket ships no discrete feature. - Record at most one such row per ticket. If you record more than one, the merge uses the LAST — so re-record a corrected full payload rather than a partial patch.
- The summary MUST start with
-
Stop. Do not call any digest or feature tool yourself — applying is the merge's job, and only the merge's, so a rejected delivery is never applied. Once this and your AC evidence are recorded you are done — the runner records the delivery and submits for review.
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 · 61 lines · 116 tokens per session scan A 5ff0b5f3143f
prepare-digest-delta is a skill published in the GitHub repository tmj-90/gaffer (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 116 tokens to every session and 943 once invoked, about $0.0006 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
bulwark-brainstorm
Role-based brainstorming with dual modes: --scoped (sequential Task tool, 5 roles) and --exploratory (Agent Teams peer debate, 4 roles). Use for feasibility assessment and idea validation.
plan-creation
Create structured implementation plans via a 4-role scrum team (Product Owner, Architect, Eng/Delivery Lead, QA/Critic) with optional Agent Teams peer debate mode.
anthropic-validator
Validates Claude Code assets (skills, hooks, agents, commands, MCP servers, plugins) against official Anthropic standards. Fetches latest docs dynamically and produces structured validation reports.
code-review
Comprehensive code review with distinct aspect based sections. Use when reviewing code, checking for security issues, finding type safety problems, auditing code quality, or when user asks to review code, PRs or changes. Three-phase workflow runs static tools, LLM judgment, and writes diagnostic log.
create-subagent
Generates single-purpose Claude Code sub-agents for use via the Task tool. Use when creating dedicated sub-agents, scaffolding agent definitions, or generating agents with diagnostics and permissions setup.
test-audit
Audit test suites for T1-T4 violations using AST analysis, mock detection, and multi-stage synthesis. Invoke when user asks to audit tests, check test quality, find mock violations, review test effectiveness, or inspect test suites for over-mocking. Triggers automatic rewrites when quality gates fail.