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 skills add OutlineDriven/outline-driven-development --skill branch-prediction-and-speculationgit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/branch-prediction-and-speculation)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/branch-prediction-and-speculation"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/branch-prediction-and-speculation/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/branch-prediction-and-speculation"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/branch-prediction-and-speculation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00049 | $0.01216 |
| Opus 5 | $0.00024 | $0.00608 |
| Sonnet 5 | $0.00010 | $0.00243 |
| Haiku 4.5 | $0.00005 | $0.00122 |
Grade A, and why
branch-prediction-and-speculation 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 3d 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Branch prediction and speculation
Contract
| Field | Bound contract |
|---|---|
| Trigger | Branchy hot code underperforms, a likely or branchless refactor needs judgment, a kernel mitigation such as retpoline or KPTI needs explaining, or code branches on secret data. |
| Authority | Read-only. The skill runs perf stat on a user-named binary, reads sysfs, and answers in chat. Nothing on disk changes, so there is nothing to roll back. No remote mutation. |
| Side effect | Chat output only. perf stat writes counters to stdout. |
| Done | The answer names the branch that mispredicts or the speculation path that leaks, gives a measured branch-misses count where a binary exists, and states the fix with the condition under which it helps. |
Inputs
- Code or hot loop (required): the source or disassembly around the branch in question.
- Binary and workload (optional): needed for a measured verdict. Without them the answer is a hypothesis.
- Threat model (optional): whether the question is performance only or also side-channel safety.
Procedure
- Explain the mechanism in one pass. The front end predicts a direction for each conditional branch, executes the predicted path, and on resolve either commits or squashes the wrong-path work and refetches from the correct address. The squash cost grows with the distance from fetch to resolve, so a deeper pipeline pays more per mispredict. Backward branches are usually loop closers and predict taken; a data-dependent forward branch with no pattern is the hard case. Done when: the user can say why a given branch is predictable or not.
- Measure before changing code. Done when: a
branch-missescount and its ratio tobranchesfor the real workload is recorded, or no binary exists and the answer is marked as unmeasured.
perf stat -e branches,branch-misses ./app
Read the ratio against the workload, not against a fixed number: a tight loop over sorted data should show a ratio near zero, while a parser over random input can sit far higher and still be at its floor. Only a ratio that drops after a change proves the change.
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.
- 3d ago First seen · 60 lines · 49 tokens per session scan A f01cc380134d
branch-prediction-and-speculation is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed 4d ago), licensed Apache-2.0. It adds 49 tokens to every session and 1,216 once invoked, about $0.0002 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 2026-09-06.
Other skills, from other repositories
audit-project
Run an iterative multi-agent code audit until critical and high findings are resolved. Use when the user says "audit my code", "find all the bugs", "deep code audit", "iterative review", or "review until clean".
debug
Hypothesis-driven debugging. Use when a test fails, a crash or exception occurs, output is wrong, or an intermittent flake has no obvious cause.
optimize
Use when asked to optimize code, speed up a path, reduce allocations, repair a regression, or profile a target. Not for remote, credential, publish, deploy, or irreversible changes.
ios-build-fix
Use when asked to run /ios-build-fix to fix a failing iOS build, regenerate an Xcode project from project.yml, or correct UI behavior. Not for a clean rebuild: use ios-build-cleanup.
strike-the-root
Use when a bug, failure, flake, regression, review finding, or ticket needs the core fixed so it cannot recur. Not for greenfield features: use tdd. Not for style-only review or typo-class one-liners.
extremely-optimize
Use when asked to run a performance campaign against a measured floor. Not for hypothesis-only analysis without mutation: use fastopt.