CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-memory

Creates and queries agent expertise profiles in AGENT-EXPERTISE.md, increments file-familiarity counters after each task, and ranks candidate agents by recency and task-area match. Use when deciding which agent should handle a file, checking who last worked on a module, recording task outcomes, or assigning work based on past performance.

100

Quality

100%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Agent Memory Protocol

Expertise File

Location: .opencastle/AGENT-EXPERTISE.md — one section per agent with Strong Areas, Weak Areas, and File Familiarity tables.

Entry format: Area | Evidence | Last Updated — e.g. Server Components | Built TAS-42 | 2026-03-15. File familiarity: - src/lib/search/ — 3 tasks.

Update Triggers

TriggerAction
First-attempt successUpdate Strong
2+ retriesUpdate Weak
File modifiedIncrement familiarity
DLQ failureAdd Weak with ref
>3 months staleMark as "stale"

Retrieval & Delegation

Query before delegating, then include a concise context block in the prompt:

grep -A5 "## Developer" .opencastle/AGENT-EXPERTISE.md

Example prompt block: Agent Context: Strong — Server Components (3 tasks); Weak — Component styling (2 retries); Familiar — src/lib/search/ (2 tasks)

Update after task completion:

# Append a Strong Area entry
printf '| %s | %s | %s |\n' "Server Components" "Built TAS-42" "$(date +%Y-%m-%d)" >> .opencastle/AGENT-EXPERTISE.md

# Increment file familiarity
awk '/src\/lib\/search\// { if (match($0, /[0-9]+/)) { n = substr($0, RSTART, RLENGTH) + 1; sub(/[0-9]+[[:space:]]*tasks?/, n " tasks") } found=1 } {print} END { if(!found) print "- `src/lib/search/` — 1 task" }' \
  .opencastle/AGENT-EXPERTISE.md > tmp && mv tmp .opencastle/AGENT-EXPERTISE.md

Workflow

  1. Before delegating: Read .opencastle/AGENT-EXPERTISE.md, check Strong/Weak areas, add concise Agent Context to the prompt.
  • Validate: the selected agent has a Strong area matching the task or no conflicting Weak entries.
  1. After task completes: Update expertise (success → Strong, 2+ retries → Weak, files → Familiarity) and append file relationships to .opencastle/KNOWLEDGE-GRAPH.md.
  • Validate: the expertise file contains the new entry and the timestamp is today's date.
  1. On DLQ failure: Add Weak Area with reference to the failure ID and link to logs.
  • Validate: failure ID and link appear in the Weak Area entry.

Pruning

Prune entries older than 6 months, remove familiarity for deleted paths, and consolidate duplicates.

  • Validate: run rg "— [0-9]+ tasks" .opencastle/AGENT-EXPERTISE.md after pruning to confirm no stale paths remain.

Knowledge Graph

File dependency graph and cross-agent relationships. See KNOWLEDGE-GRAPH.md for entity types, templates, triggers, and queries.

Repository
monkilabs/opencastle
Last updated
Created

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.