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 t1djani/servo --skill gather-contextgit clone --depth 1 https://github.com/t1djani/servoWrote 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/t1djani/servo/gather-context)<a href="https://agentmods.dev/skills/t1djani/servo/gather-context"><img src="https://agentmods.dev/badge/skills/t1djani/servo/gather-context/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/t1djani/servo/gather-context"><img src="https://agentmods.dev/badge/skills/t1djani/servo/gather-context.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 | $0.00052 | $0.00557 |
| Opus 5.5 | $0.00021 | $0.00223 |
| Sonnet 5.5 | $0.00010 | $0.00111 |
| Haiku 4.5 | $0.00005 | $0.00056 |
Grade A, and why
gather-context 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 1mo 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 — 27 lines — stays where its author put it; the contents beside it link to each section on GitHub.
gather-context
Load the ground truth a task needs before touching it — and again whenever the work has moved the ground. Context decays over a long session and drifts as the work changes state; the slice of truth that matters before writing a plan is not the same one that mattered before the brainstorm. So this is a recurring ritual, not a one-time setup.
The failure this prevents: an agent re-deriving facts that already exist (in the manifest's oracles), or asking the human to re-feed context the project already records.
Procedure
-
Read the manifest.
.servo/manifest.yamllists the project'soracles(where ground truth lives) andexperts(the domains). If there is no manifest, runscan-projectfirst. -
Pull only the relevant slices. For the task at hand, decide which oracles matter (scope of the work, invariants it could break, prior art that already exists, known footguns) and read those sources — and only those. Do not ingest everything; pull the slice the task needs.
-
Write a short context brief. A few lines: what the task is, what it must not break (invariants), what already exists (prior art), which footguns apply, which experts are relevant. Cite the oracle each fact came from.
-
Confirm in one batch — not one question at a time. Where the brief has gaps or assumptions, ask the human all the open questions at once, with proposed answers pre-filled from the context. The human is the CEO/client seat here: they own the intent. Their answers are input to pressure-test, not orders to execute blindly.
-
Carry the brief forward — it is the cache. The brief is read ONCE here and then consumed by the downstream phases (the expert panel, the gates). Hand subagents the brief instead of having each one re-read the raw oracles; a subagent only goes back to a raw source when it genuinely needs to go deeper than the brief. Reading the same vault notes and specs eleven times, once per subagent, is the main avoidable cost in the flow. At the next phase boundary, re-run this skill on the slice that now matters.
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.
- 1mo ago First seen · 27 lines · 52 tokens per session scan A cd231a0b9551
gather-context is a skill published in the GitHub repository t1djani/servo (1 stars, last pushed 4mo ago), licensed MIT. It adds 52 tokens to every session and 557 once invoked, about $0.0002 per session on Opus 5.5. 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
agent-memory
Repo-local file memory under .scaffolding/ (shared/agent/conversation tiers). TRIGGER when: reading or writing .scaffolding/ memory files, or choosing which file tier an insight belongs in. SKIP: vector memory / MCP search and store (use semantic-memory-mcp); distilling conversations (use distill).
semantic-memory-mcp
Search, recall and store using the memory MCP tools (searchcontext, semanticsearch, semanticrecall, semanticstore, notes). TRIGGER when: searching prior insights mid-task, or deciding whether and how to store an insight or note. SKIP: repo file tiers under .scaffolding/ (use agent-memory); scoring /learn candidates…
distill
Knowledge distillation: candidate extraction, confidence scoring, tier routing. Invoked by /learn and the memory-ingest Stop hook, not loaded proactively by agents. TRIGGER when: /learn or the Stop hook asks to distill, or scoring a candidate. SKIP: file memory writes (use agent-memory); vector store (use…
save
Save durable decisions, findings, plans, and implementation knowledge from the visible Codex conversation into a new or existing NeatContext context. Use only when the user explicitly invokes this skill or asks to preserve the current conversation as reusable context.
import
Import a self-contained NeatContext context bundle shared by another person, or reconcile a newer copy of a context already on this machine, leaving the source bundle unchanged. Use only when the user explicitly invokes this skill or asks to import a NeatContext bundle.
quiz-me
Quiz the user on root cause and intended fix BEFORE writing any non-trivial code, then verify comprehension of every change after. Use when the user asks for a bug fix, refactor, or feature and wants to stay technically sharp instead of vibe-coding. Also use when the user says "quiz me", "don't let me vibe code", or…