Self Configuration

What it does

When your agent uses it

• User asks to self-modify
• Tune summarization/compaction
• Change identity/system instructions
• Adjust model settings

Example prompts

• “Use the self-configuration skill to configure Letta agents' own runtime behavior, including model, context window, system prompt, reasoning…”
• “/self-configuration”

Requirements

• Python 3
• Node.js
• A credential in LETTA_API_KEY

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1Save or inspect the current agent/conversation configuration.
2. 2Patch the current conversation, not persistent agent defaults.
3. 3Verify the response or re-fetch the conversation to confirm the config changed.
4. 4Run a tiny low-risk runtime test in the same conversation.
5. 5If the runtime test fails, revert the conversation to the saved known-good model/settings.

What it can do on your machine

Read from SKILL.md and the folder at commit 6785511. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 2 files in scripts/ (TypeScript), which the agent can run.

  Shell commands in SKILL.md call:

  • curl
  • jq
  • npx

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use curl and npx, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • LETTA_API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Self Configuration loads about 2.4k tokens when it runs, and up to ~5.8k if it reads all its reference files. Until then it costs about 93 tokens; SKILL.md has 786 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~93

When it runs · the whole SKILL.md, loaded when a task matches

~2.4k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~5.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from letta-ai/skills at commit 6785511, republished under its MIT licence (© letta-ai). 786 words, ~2,381 tokens.

name

self-configuration

description

Configures Letta agents' own runtime behavior, including model, context window, system prompt, reasoning, conversation overrides, compaction settings, and compaction prompts. Use when an agent or user asks to self-modify, tune summarization/compaction, change identity/system instructions, adjust model settings, or test conversation-scoped overrides.

license

MIT

BASE_URL="${LETTA_BASE_URL:-https://api.letta.com}"
: "${LETTA_API_KEY:?Set LETTA_API_KEY}"
: "${AGENT_ID:?Set AGENT_ID}"

curl -sS "$BASE_URL/v1/agents/$AGENT_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" | \
  jq '{id, model, context_window_limit, llm_config, model_settings, compaction_settings, system_chars: (.system | length)}'

curl -sS -X PATCH "$BASE_URL/v1/agents/$AGENT_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"context_window_limit": 64000}'

: "${CONVERSATION_ID:?Set CONVERSATION_ID}"

curl -sS -X PATCH "$BASE_URL/v1/conversations/$CONVERSATION_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"context_window_limit": 64000}'

curl -sS -X PATCH "$BASE_URL/v1/agents/$AGENT_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.2",
    "context_window_limit": 272000,
    "model_settings": {
      "provider_type": "openai",
      "parallel_tool_calls": true,
      "reasoning": { "reasoning_effort": "medium" },
      "max_output_tokens": 128000
    }
  }'

: "${CONVERSATION_ID:?Set CONVERSATION_ID}"

curl -sS -X PATCH "$BASE_URL/v1/conversations/$CONVERSATION_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.2",
    "context_window_limit": 64000,
    "model_settings": {
      "provider_type": "openai",
      "parallel_tool_calls": true,
      "reasoning": { "reasoning_effort": "low" }
    }
  }'

curl -sS "$BASE_URL/v1/agents/$AGENT_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" | jq -r '.system'

curl -sS -X PATCH "$BASE_URL/v1/agents/$AGENT_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"system": "<FULL replacement system prompt. Preserve important existing instructions.>"}'

npx tsx <SKILL_DIR>/scripts/update-agent-settings.ts \
  --target agent \
  --context-window-limit 64000 \
  --dry-run

npx tsx <SKILL_DIR>/scripts/update-compaction-prompt.ts \
  --prompt-file /tmp/compaction-prompt.txt \
  --mode self_compact_sliding_window \
  --clip-chars 50000 \
  --dry-run

curl -sS "$BASE_URL/v1/agents/$AGENT_ID" \
  -H "Authorization: Bearer $LETTA_API_KEY" | \
  jq '{id, model, context_window_limit, llm_config_context_window: .llm_config.context_window, model_settings, compaction_settings, system_chars: (.system | length)}'

© letta-ai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (scripts, references) in letta/self-configuration of letta-ai/skills.

• SKILL.md
• LICENSE
• references/api-patch-examples.md
• references/compaction-prompt-patterns.md
• references/model-settings.md
• scripts/update-agent-settings.ts
• scripts/update-compaction-prompt.ts

Open the folder on GitHubat commit 6785511