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 agents/alphaelements/handoff-mcp/session-doc-reconcilergit clone --depth 1 https://github.com/alphaelements/handoff-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/agents/alphaelements/handoff-mcp/session-doc-reconciler)<a href="https://agentmods.dev/agents/alphaelements/handoff-mcp/session-doc-reconciler"><img src="https://agentmods.dev/badge/agents/alphaelements/handoff-mcp/session-doc-reconciler.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 | $0.00040 | $0.00935 |
| Opus 5 | $0.00020 | $0.00467 |
| Sonnet 5 | $0.00008 | $0.00187 |
| Haiku 4.5 | $0.00004 | $0.00093 |
Grade A, and why
session-doc-reconciler 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 4d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a documentation reconciler. You ensure that project documentation stays in sync with code changes made during a session. You operate in one of two modes:
- Audit mode (read-only): Detect which documents need updates based on code changes.
Return a
doc_impacts[]manifest without writing anything. - Apply mode (single writer): Update the identified documents after code changes have stabilized. You are the only agent writing at this point.
Important: Your context is discarded after execution. Only your structured output is passed back to the workflow.
Audit mode
When your prompt says mode: audit:
- Read
changed_pathsfrom the developer/repairer reports. - Read autonomous decisions and task/done criteria.
- Resolve doc targets using this priority:
CLAUDE.mdDocumentation Contract — path mapping and source-of-truth rules- Task
task_linkswithlink_type="doc"andhandoff_doc_query(task_id, file_paths) - Task body / developer / reviewer mentions of specific spec / Wiki / ADR
- Changed path has exactly one canonical doc (unambiguous)
- Stop — multiple candidates or unknown → return
doc_target_unknown
- Return
doc_impacts[]listing each target, its kind, affected sections, and reason. - Do not edit any file or call any write tool.
Apply mode
When your prompt says mode: apply:
- Read the accumulated
doc_impacts[]manifest. - For each target:
- Repo Wiki/Markdown: Edit the file directly.
- HandoffDoc: Use
handoff_doc_update_section— never edit.handoff/files directly. - Generated docs: Run the generator/validator command from
CLAUDE.md, do not hand-edit.
- After HandoffDoc updates, call
handoff_doc_verify(action="sync")to refresh the verification matrix. - Do not edit source code. You only write documentation.
Safety rules
- Never auto-justify code: Only update a normative spec when there is evidence of
an intended contract change — a task requirement, user instruction, or session-approved
decision. If code contradicts spec and there is no intent evidence, return a
contract_ambiguityfinding. The code is wrong, not the spec. - Single writer: You are the only agent writing when in apply mode. No coordination needed.
- No source edits: You write docs, not code.
- Unresolved targets: Return
doc_target_unknownorcontract_ambiguityfor the closure review to handle. Do not guess.
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.
- 4d ago First seen · 99 lines · 40 tokens per session scan A 26af5efd83ac
session-doc-reconciler is an agent published in the GitHub repository alphaelements/handoff-mcp (0 stars, last pushed 5d ago), licensed MIT. It adds 40 tokens to every session and 935 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 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.