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 edg-l/engram-mcp --skill handoffgit clone --depth 1 https://github.com/edg-l/engram-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/edg-l/engram-mcp/handoff)<a href="https://agentmods.dev/skills/edg-l/engram-mcp/handoff"><img src="https://agentmods.dev/badge/skills/edg-l/engram-mcp/handoff/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/edg-l/engram-mcp/handoff"><img src="https://agentmods.dev/badge/skills/edg-l/engram-mcp/handoff.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 | $0.00047 | $0.00973 |
| Opus 5 | $0.00023 | $0.00487 |
| Sonnet 5 | $0.00009 | $0.00195 |
| Haiku 4.5 | $0.00005 | $0.00097 |
Grade A, and why
handoff 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Capture session state into Engram via the mcp__engram__handoff_create MCP tool. Engram is now the canonical store; no markdown file is written.
Steps
-
Gather working state (skip if not in a git repo). Run in parallel:
git branch --show-currentgit log --oneline -5git status --shortgit diff --statgh pr view --json url,number,state 2>/dev/null
-
Build the handoff payload. Map session content to these sections.
summaryis required; everything else is optional. Omit sections with nothing real to say; do not pad.summary(string): 1–3 sentences naming what was worked on.decisions(string array): architectural / design choices made this session, one per item, with rationale.todos(string array): remaining tasks. Specific: file path, function name, the next concrete action.blockers(string array): approaches tried that did not work, with why each failed. So the next agent does not retry them.mental_model(string): prose paragraph describing how the system / problem fits together right now. Use when the next agent needs the model in their head.next_steps(string array): ordered/prioritized actions, closely related totodos. Usenext_stepsfor "what to do next"; usetodosfor "what's still open".notes(string, optional): open questions for the user, env quirks, pre-existing failures, anything that doesn't fit the other sections.
-
Detect
continues_from. If theread-handoffsskill ormcp__engram__handoff_resumeran earlier this session and returned alatest_handoff_id, pass that id ascontinues_from. Otherwise omit. -
Sensitive data filter. Before calling the tool, scrub: API tokens, passwords, private URLs, customer data, internal hostnames. If unsure, ask the user.
-
Call
mcp__engram__handoff_createwith the payload. Defaults:pinned: true,importance: 0.85,auto_link: true. Branch is auto-detected from git; pass an explicitbranchonly if you want to override. -
Report to the user. Print the new handoff id, the auto-linked memory count (decisions/patterns/debug memories Engram associated by similarity), and the
continues_fromlink if one was set. Ask if any section should be revised; if so, the user can dictate edits and you can callmcp__engram__memory_updateon the just-created handoff. -
Escalate single insights worth keeping outside the handoff. Only do this for content that should surface in
memory_contextqueries on unrelated future tasks. Callmcp__engram__memory_storeseparately for:type: decision— architectural choice with rationale (importance 0.7)type: pattern— recurring gotcha / convention discovered (importance 0.7)type: debug— root cause of a tricky bug (importance 0.6) The handoff already captures session-level context; only escalate insights with cross-session value.
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 · 53 lines · 47 tokens per session scan A b84bac0695fe
handoff is a skill published in the GitHub repository edg-l/engram-mcp (15 stars, last pushed 8d ago), licensed Apache-2.0. It adds 47 tokens to every session and 973 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
engrama-onboard
Build a personalized Engrama memory graph for any user through a conversational interview. Use this skill whenever someone says "set up Engrama", "create my graph", "onboard", "build my profile", "configure Engrama for me", or asks how to start using Engrama. Also use when the user mentions they have a different role…
memory-review
Memory hygiene audit -- finds stale facts, contradictions, low-confidence entries, and consolidation candidates in Pensyve memory. Use periodically to maintain memory quality.
session-memory
End-of-session memory capture -- classifies session signals using a tiered taxonomy and stores confirmed items via Pensyve. Use when ending a work session or when the user wants to capture what was learned.
context-loader
Session-start context loading -- loads historical decisions, issues, and patterns from Pensyve to provide cross-session continuity. Use at session start or when switching context.
memory-informed-refactor
Pre-refactor context briefing -- loads relevant prior decisions, failures, and pitfalls from Pensyve memory before refactoring a module. Use before any refactor to avoid repeating past mistakes.
memory-informed-design
Architecture/design decisions with working memory -- before recommending, recall prior decisions and tradeoffs; when a decision is made, capture it immediately. Use for any substantive design question.