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/msolecki/founder-os/context-loadnpx skills add msolecki/founder-os --skill context-loadgit clone --depth 1 https://github.com/msolecki/founder-osWhat 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 | $0.00028 | $0.02053 |
| Opus 5 | $0.00014 | $0.01026 |
| Sonnet 5 | $0.00006 | $0.00411 |
| Haiku 4.5 | $0.00003 | $0.00205 |
Grade A, and why
context-load 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 2d 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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Load
Agents have amnesia. Every cadence starts cold, and a cold agent that opines anyway is doing the one thing house rule 1 forbids: guessing, which is a service the founder can already perform for free.
The dangerous state is not missing context. Missing context is loud — the file is empty and the agent stops. The dangerous state is stale context that reads as fresh: a runway figure from a close that happened seven weeks ago, quoted today with full confidence and no date attached. This skill exists to put the date on the number.
When to use
At the start of every cadence — daily brief, week plan, weekly review, monthly close, quarterly planning — and at the start of any session where an agent is about to make a claim about the business.
Also whenever a session has been running long enough that a number was quoted earlier in the conversation. Re-read the file. "As we established above" is memory, and memory is exactly what this package does not trust.
Inputs
Three files, always, in this order. The order is the point:
charter.md— what this business is. Read first because it tells you what to ignore. Everything after this is filtered through it.goals.md— this quarter's bets and their kill conditions. What is currently being attempted, and what would end it.metrics.md— the numbers, and the date they were last true.
Then, and only then, the one file your decision owns.
Steps
- Resolve which business, before any file is opened. If
~/.founder-os/businesses.yamldoes not exist, this step is free: the workspace isFOUNDER_OS_HOMEor./founder-os/, as it always was. If it exists, apply the precedence inreferences/multi-business.md: a slug named in the invocation or the session wins; elseFOUNDER_OS_HOME(mapped back to its slug for display); else the registry'sdefault:. With more than one active business and no signal, ask — one question, one word of answer. Guessing which company a question is about is not a time saving; it is advice filed under the wrong business, quoted back from the wrongmetrics.mdfor a quarter. The resolved slug opens the context line so the founder sees which company is being advised before they read the advice.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 151 lines · 28 tokens per session scan A 098744277cdd
context-load is a skill published in the GitHub repository msolecki/founder-os (4 stars, last pushed 8d ago), licensed MIT. It adds 28 tokens to every session and 2,053 once invoked, about $0.0001 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…