Agent Journal

Give AI agents a persistent journal backed by BlueColumn semantic memory. Use when an agent should log daily observations, decisions, user preferences, or no...

Security Scan

The skill's behavior (sending agent data to BlueColumn endpoints) is coherent with a journaling purpose, but the SKILL.md requires a BlueColumn API key while the registry metadata declares no credentials and references a missing TOOLS.md — this mismatch and missing provenance are concerning.

Purpose & Capability

The skill's stated purpose (persisting agent notes to BlueColumn) aligns with the HTTP endpoints documented in SKILL.md and references/api.md. However, the SKILL.md explicitly says a BlueColumn API key (bc_live_*) is required, yet the registry metadata lists no required environment variables or primary credential. That inconsistency means the skill's declared needs do not match what it actually requires to function.

Instruction Scope

Runtime instructions are narrowly scoped to three BlueColumn endpoints (/agent-note, /agent-remember, /agent-recall) — all expected for a journal. They do not instruct reading arbitrary local files or other secrets. However, the SKILL.md tells the agent to 'Read TOOLS.md for the BlueColumn API key' but no TOOLS.md is included in the package, which is ambiguous and could lead to ad-hoc secret prompting or misconfiguration.

Install Mechanism

This is instruction-only with no install spec and no code files, so nothing will be written to disk by an installer. That is the lowest-risk install model.

Credentials

The skill needs a sensitive credential (bc_live_* API key) to contact an external service, but the manifest does not declare any required env var or primary credential. That omission is disproportionate/unhelpful: users and the platform cannot see at install-time that a secret is required. Additionally, the skill will send user data to BlueColumn's Supabase backend (external network egress), which is proportionate to journaling but requires explicit user consent and a privacy review.

Persistence & Privilege

always is false (normal). The skill does not request elevated or persistent system privileges and does not claim to modify other skills or system settings.

What to consider before installing

Before installing, ask the author to correct the manifest so the BlueColumn API key is declared (e.g., add a required env var or primary credential) and to include or point to the missing TOOLS.md. Confirm the exact environment variable name the agent expects and how keys are stored/rotated. Verify you trust BlueColumn (review bluecolumn.ai privacy and data retention), since journal entries will be sent to a Supabase endpoint (https://xkjkwqbfvkswwdmbtndo.supabase.co/functions/v1). If you proceed, avoid sending highly sensitive data, create a dedicated API key with minimal scope and rotation, and consider requiring explicit user invocation (disable autonomous invocation) or testing in a sandbox account first.

Agent Journal Skill

Persistent cross-session journal for AI agents. Store observations, preferences, and decisions — recall them anytime.

Setup

Read TOOLS.md for the BlueColumn API key (bc_live_*). Keys are generated at bluecolumn.ai/dashboard. Store securely — never log or expose them.

Base URL: https://xkjkwqbfvkswwdmbtndo.supabase.co/functions/v1 (BlueColumn's official backend — bluecolumn.ai runs on Supabase Edge Functions)

Log an Observation (agent-note)

For lightweight agent observations — preferences, patterns, decisions:

curl -X POST .../agent-note \
  -H "Authorization: Bearer <key>" \
  -d '{"text": "User prefers bullet points over paragraphs", "tags": ["preference", "style"]}'

Log a Detailed Entry (agent-remember)

For richer entries with auto-extracted metadata:

curl -X POST .../agent-remember \
  -H "Authorization: Bearer <key>" \
  -d '{"text": "Session summary: user is building a SaaS product in Phoenix. Decided on PostgreSQL. Main concern is scalability.", "title": "Session Log 2026-04-14"}'

Recall Journal Entries

curl -X POST .../agent-recall \
  -H "Authorization: Bearer <key>" \
  -d '{"q": "what does the user prefer for formatting?"}'

When to Use Each

Situation	Endpoint
Quick preference or observation	/agent-note
End-of-session summary	/agent-remember
Retrieving past context	/agent-recall

Journaling Workflow

Start of session: Recall relevant context:

• "What do I know about this user?"
• "What were we working on last time?"

During session: Note important observations via /agent-note

End of session: Summarize and store via /agent-remember

Tag Suggestions

• preference — user style/format preferences
• decision — architectural or product decisions
• context — background info about user/project
• followup — things to revisit

See references/api.md for full API reference.