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 skills/obto-inc/platform/obto-memorynpx skills add obto-inc/platform --skill obto-memorygit clone --depth 1 https://github.com/obto-inc/platformWrote 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/skills/obto-inc/platform/obto-memory)<a href="https://agentmods.dev/skills/obto-inc/platform/obto-memory"><img src="https://agentmods.dev/badge/skills/obto-inc/platform/obto-memory.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 | $0.00075 | $0.00560 |
| Opus 5 | $0.00037 | $0.00280 |
| Sonnet 5 | $0.00015 | $0.00112 |
| Haiku 4.5 | $0.00007 | $0.00056 |
Grade A, and why
obto-memory 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 5d 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 — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OBTO Memory Discipline (obto_remember / obto_recall)
Memories are backed by Hindsight (vector-embedded) and survive session resets and pod reboots. Used well, they remove the "agent forgets everything between conversations" blocker for autonomous work.
Recall before work
Call obto_recall before making decisions or starting work on anything you may have notes about — app architecture, known quirks, prior decisions. It is fail-soft: if the store is unavailable it returns {ok:true, returned:0} rather than blocking.
Two scoping modes — pick one per call
App-scoped (default for interactive/frontier-model agents): pass appName + domain. Memory is long-lived, scoped to (user, app, domain). Use for durable knowledge ABOUT an app: architecture, conventions, gotchas, preferences.
Conversation-scoped (for API-built small-model agents): pass conversationId (a UUID you generate, stable for the conversation). Memory becomes the conversation's external working memory — write active app/domain and recent decisions at the start of each turn, read them back to drive tool calls. This is the platform's answer to maintaining context under the stateless contract (see obto://guide/quickstart Rule 0.5).
Write discipline
- Supply a stable
keyfor facts that should be updated in place — same (key, scope) overwrites. Omitkeyonly for append-style notes. - Write facts worth re-reading: decisions and why, non-obvious constraints, verified procedures. Do not store transient state (use
obto_set_propertyfor config keys), secrets, or anything needing strong consistency. - Keep entries self-contained — recall is similarity-based, so an entry should make sense without surrounding context. Convert relative dates to absolute.
- Verify the returned
mode/projectKeymatches the scope you intended.
Effective autonomous pattern
- Turn start:
obto_recall(query=<task topic>, appName, domain). - Do the work.
- Turn end:
obto_rememberanything a future session would need — keyed, scoped, dated.
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.
- 5d ago First seen · 38 lines · 75 tokens per session scan A e06c4fe2e08c
obto-memory is a skill published in the GitHub repository obto-inc/platform (1 stars, last pushed 2d ago), licensed Apache-2.0. It adds 75 tokens to every session and 560 once invoked, about $0.0004 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 skills, from other repositories
memory-audit-pattern-extraction
模式提取与失效解药分析。当发现多条记忆在讲同一个教训，或发现自己在一而再再而三地犯同样的错误时使用。.
init-workspace-documentation
Skill "init-workspace-documentation" from griddynamics/rosetta, covering agent memory.md, agent memory, preventive rules, what worked and what failed.
memorix-mini-skills
Use when durable project knowledge, gotchas, workflows, or repeated fixes should become reusable agent guidance instead of ordinary memory.
memorix-sessions
Use when resuming work, preparing handoff context, binding an HTTP control-plane project, or deciding whether sessionstart is useful.
frontmcp-testing
Use for anything about testing FrontMCP servers: writing or running unit, integration, and E2E tests and reaching the 95%+ coverage bar. Covers Jest setup and coverage gating; unit-testing a ToolContext execute() with mock context, inputs, and Zod schema validation; testing resources and prompts; in-memory testing via…
cortex-profile
View and manage your cognitive profile — how you think, work patterns, blind spots, and cross-domain connections. Use when the user says 'show my profile', 'how do I work', 'what are my patterns', 'cognitive style', 'blind spots', 'methodology', or at the start of a session to load context. Also use 'rebuild profile'…