Performance Profiling

What it does

When your agent uses it

• Asked to improve speed
• Fix memory leaks
• Find bottlenecks
• Optimize a slow function

Example prompts

• “/performance-profiling”

Requirements

• Python 3
• Compatibility (from SKILL.md): Requires python 3.10+ or node 18+

Workflow steps

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

1. 1Establish a baseline. Before changing anything, measure the current performance
2. 2Read the profiler output. Identify the top 3 functions by cumulative time. Do not optimize anything that is NOT in the top 3 — this is…
3. 3Form ONE hypothesis for why the top bottleneck is slow. Common causes, in order of frequency
4. 4Apply ONE change. The smallest possible change that addresses the hypothesis. Do not refactor the entire module.
5. 5Measure again. Compare against the baseline. Report the improvement as a percentage.
6. 6Repeat. If there is still a performance gap, return to step 2 with the updated profile.

What it can do on your machine

Read from SKILL.md and the folder at commit 2b21998. 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

  No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

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

• Network

  No URLs in SKILL.md.

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

• Credentials

  Names no API keys, tokens, secrets or passwords.

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

• Compatibility

  Requires python 3.10+ or node 18+

  From compatibility in the SKILL.md frontmatter.

Context cost

Performance Profiling loads about 804 tokens when it runs. Until then it costs about 48 tokens; SKILL.md has 298 words of instructions outside code blocks.

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

~48

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

~804

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); files beside SKILL.md are not scanned.

SKILL.md

The full file from agulli/atlas-agents at commit 2b21998, republished under its MIT licence (© agulli). 298 words, ~804 tokens.

name

performance-profiling

description

Profile and optimize application performance. Use when asked to improve speed, reduce latency, fix memory leaks, find bottlenecks, or optimize a slow function or endpoint.

compatibility

Requires python 3.10+ or node 18+

license

MIT

# Before: queries once per user
for user in users:
    user.orders = db.query(f"SELECT * FROM orders WHERE user_id={user.id}")

# After: one query for all users
user_ids = [u.id for u in users]
orders = db.query("SELECT * FROM orders WHERE user_id = ANY(%s)", [user_ids])
orders_by_user = defaultdict(list)
for o in orders: orders_by_user[o.user_id].append(o)
for user in users: user.orders = orders_by_user[user.id]

from functools import lru_cache

@lru_cache(maxsize=512)
def expensive_computation(input_id: int) -> dict:
    ...

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

Files

Just SKILL.md in ch09_agent_skills/skills/performance-profiling of agulli/atlas-agents.

Open the folder on GitHubat commit 2b21998