infernoflow
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| INFERNOFLOW_AGENT | No | Override the automatically detected agent/writer name stamped on memory entries (e.g. claude, cursor, copilot, memory-keeper, hook, human). | |
| INFERNOFLOW_PROJECT_DIR | No | Project folder whose .ai-memory/ the server reads and writes. Defaults to the current directory. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| amp_readC | AMP: read session memory entries with optional filters. |
| amp_writeA | AMP: log a new entry. Required: type + msg (one sentence). Optional: file, line, tags, detail. Call it without being asked when: the user is stuck ('!!', 'not working', 'same error', 'retry' → type attempt: what was tried and how it failed); the user asks you to remember something (save the fact in your own words); a decision is made (type decision: X over Y because Z); you learn something non-obvious (type gotcha, with file/line). These apply in any language the user writes in. Don't log what can be re-derived from the code. Use 'detail' for a rich multi-paragraph body (repro steps, code, full reasoning, or a session snapshot) — it's stored in a sidecar and loaded on demand, so it never bloats the always-on memory index. |
| amp_searchC | AMP: search entries by keyword. Optional type filter. |
| amp_bookmarkA | AMP: drop a named session bookmark — a resume point. Required: label (short name). Optional: note. If note is OMITTED, the current session transcript is auto-captured as the bookmark's context (the 'save everything here' resume point). Use when the user says 'bookmark this' / 'mark this point', or before a risky change / when the context window is filling up, so the exact state can be recalled later and appears in the next session's handoff. Bookmarks are never auto-pruned. |
| amp_resumeA | AMP: 'where were we?' in one call — the latest resume point with its note, open dead ends (don't repeat them), recent decisions/notes, uncommitted changes, and which memory store this is. Call it at the start of work in a session. Optional: file (rank entries about that file first). |
| amp_handoffB | AMP: generate the handoff document for the next AI session. format=markdown|json (default: markdown). |
| amp_healthA | AMP: get the session health score (0-100, A-F grade). |
| infernoflow_statusC | Show project memory + contract health at a glance. |
| infernoflow_git_driftB | Detect which capabilities may be affected by recent code changes — useful when memory needs branch-aware revalidation. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Each tool targets a fairly distinct operation: write/read/search/bookmark/resume/handoff cover the memory lifecycle, while git_drift and status cover project-level health. The main overlap is between amp_health (session health score) and infernoflow_status (project memory + contract health 'at a glance'), and amp_resume vs amp_read are close since resume is a composite view over the same entries. Descriptions do enough to steer selection, but the health/status pair could be confused.
All nine names follow a snake_case verb_noun pattern (amp_bookmark, amp_write, infernoflow_git_drift), which is predictable and readable. The only deviation is the split between the amp_ and infernoflow_ prefixes, which implies two namespaces inside a single server and slightly muddies grouping. Internally each prefix is consistent, so this is a minor deviation rather than a mixed-convention problem.
Nine tools is well within the ideal 3-15 range for a session-memory server. Each tool maps to a concrete step in the workflow (capture, retrieve, summarize, health-check) with no filler entries. Nothing appears redundant enough to be pruned.
The capture-and-retrieve surface is solid: write, read, search, bookmark, resume, handoff, plus health and drift diagnostics. However there is no update or delete/forget operation, so a wrong or stale memory entry cannot be corrected or removed — only appended to, which is a notable gap for a memory store. Listing or pruning bookmarks is also absent.