Borrowing it
Nothing to install: this file belongs to fulminate-io/knowledge-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/fulminate-io/knowledge-mcp/main/.claude/skills/prefill/SKILL.mdgit clone --depth 1 https://github.com/fulminate-io/knowledge-mcpWrote 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/fulminate-io/knowledge-mcp/prefill)<a href="https://agentmods.dev/skills/fulminate-io/knowledge-mcp/prefill"><img src="https://agentmods.dev/badge/skills/fulminate-io/knowledge-mcp/prefill.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 | $0.00073 | $0.01589 |
| Opus 5 | $0.00036 | $0.00794 |
| Sonnet 5 | $0.00015 | $0.00318 |
| Haiku 4.5 | $0.00007 | $0.00159 |
Grade A, and why
prefill 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 today.
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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PREFILL — the implementer's preloaded context
What a prefill is for
The prefill exists so the implementer implements instead of asking. It is everything an engineer new to this part of the codebase would need to build the ticket correctly on the first try: where the change lands, what to reuse, what it must not break, how it will be tested, and where it lands. It is a plan ROOT with no phases and no steps, carrying the goal, the tree, the reads and the citations block and no section text at all; each numbered section below is its own section node on a positioned contains edge, written alone and read alone.
The citation rule
Every file, line, symbol, count and claim in a prefill carries the command that resolved it and was opened this session at the tree the prefill names. A citation without its command is an assertion, and an assertion in a prefill is a T1 in review. Counts are tool output pasted beside the command, labeled as floors; a hand-written "all N sites" does not exist in a prefill.
The shape
The root is an INDEX, not a document. Its sections hang off it as section
nodes on contains edges carrying a zero-based position, in the order below;
the root's tree lists them in that order with each one's size, so a reader
chooses its pages rather than loading the whole plan. Changing a section is
ONE write to that section node: no other section and not the root is
re-sent, and a planner that re-writes the whole prefill to fix a sentence has
made the mistake this shape exists to prevent.
A reviewer attaches an ANNOTATION to the section it concerns, never to the
root: correct, finding with its tier, or needed change carrying the
exact replacement text in the annotation's body. The annotation's kind and
tier ride the edge that joins it to the section as well as the annotation
node, so a reader ranks a section's annotations from an edge-metadata
traverse without opening one. The section's own body is never edited by the
reviewer. A needed change is applied by exact replacement of the quoted text,
never by retyping the section.
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.
- today First seen · 121 lines · 73 tokens per session scan A 338a6f9384de
prefill is a skill published in the GitHub repository fulminate-io/knowledge-mcp (1 stars, last pushed yesterday), licensed Apache-2.0. It adds 73 tokens to every session and 1,589 once invoked, about $0.0004 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 2026-09-07.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…