chore(docs): improve ai instructions by webern · Pull Request #1582 · DataDog/saluki

webern · 2026-05-05T11:33:38Z

Summary

Claude was routinely opting not to read AGENTS.md. I found that even when I added a .claude/CLAUDE.md which tells it to read AGENTS.md it still skipped that step.

I added a session start hook that more-or-less forces Claude to actually read the AGENTS.md file and I added a section about our building and testing commands to AGENTS.md

I also added some information to help it understand what commands we need to use to check our work. It was routinely reporting being done with work without running fmt or the linter. I think it's still going to skip those steps but at least now we can scold it and say "Did you run the linting steps?" and it will have those instructions available to say "Oh sorry, blah blah, here i go with make fmt"

Note: some of it could get stale quickly. We should actually create a skill and just run it occasionally to look for stale instructions or paths. But keeping this PR quick and small.

Change Type

Non-functional (chore, refactoring, docs)

How did you test this PR?

Before

On main, when I start a session and say "run the integration tests", it looks around and chooses make test-integration-quick to save time, i.e. not building the containers. That's most likely not what you want:

⏺ Let me check how integration tests are run in this project.

⏺ Bash(ls /my/weird/path//saluki-main/)
  ⎿  AGENTS.md
     bin
     bun.lock
     Cargo.lock
     Cargo.toml
     ci
     clippy.toml
     deny.toml
     docker
     docs
     lib
     LICENSE
     LICENSE-3rdparty.csv
     license-tool.toml
     Makefile
     NOTICE
     package.json
     README.md
     renovate.json5
     repository.datadog.yml
     rust-toolchain.toml
     rustfmt.toml
     SECURITY.md
     static-analysis.datadog.yml
     target
     test
     tomlfmt.toml
     tooling

⏺ Bash(grep -i "integrat\|test-correct\|test-integration" /my/weird/path//saluki-main/Makefile | head -40)
  ⎿  .PHONY: test-correctness
     test-correctness: build-panoramic
     test-correctness: ## Runs the complete correctness suite (all test cases in parallel)
     .PHONY: test-correctness-case
     test-correctness-case: build-panoramic
     test-correctness-case: ## Runs a single correctness test case by name (usage: make test-correctness-case CASE=dsd-plain)
     build-panoramic: ## Builds the panoramic binary (ADP integration test runner)
     .PHONY: test-integration
     test-integration: build-panoramic build-datadog-agent-image
     test-integration: ## Runs all ADP integration tests
        @echo "[*] Running ADP integration tests..."
        @target/release/panoramic run -d $(shell pwd)/test/integration/cases $(if $(PANORAMIC_LOG_DIR),-l $(PANORAMIC_LOG_DIR))
     .PHONY: test-integration-quick
     test-integration-quick: build-panoramic
     test-integration-quick: ## Runs ADP integration tests (assumes images already built)
        @echo "[*] Running ADP integration tests (quick mode)..."
        @target/release/panoramic run -d $(shell pwd)/test/integration/cases $(if $(PANORAMIC_LOG_DIR),-l $(PANORAMIC_LOG_DIR))
     .PHONY: list-integration-tests
     list-integration-tests: build-panoramic
     list-integration-tests: ## Lists available ADP integration tests
        @target/release/panoramic list -d $(shell pwd)/test/integration/cases

⏺ Bash(make test-integration-quick 2>&1)

After

It reads AGENTS.md and understands what to do:

⏺ Let me read AGENTS.md first as instructed.

  Searched for 1 pattern, read 1 file (ctrl+o to expand)

⏺ Let me check the Makefile to confirm the integration test targets.

  Searched for 1 pattern, read 1 file (ctrl+o to expand)

⏺ The integration suite uses make test-integration, which builds images first. Running it now — this will take a while since it builds the Datadog Agent image.

⏺ Bash(make test-integration 2>&1)
  ⎿  Running in the background (↓ to manage)
  ⎿  (timeout 10m)

References

N/A

Add build and test instructions to AGENTS.md to improve Claude Code performance.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 875c30a480

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

pr-commenter · 2026-05-05T11:43:56Z

Binary Size Analysis (Agent Data Plane)

Target: e8a875d (baseline) vs 803ddec (comparison) diff
Analysis Type: Stripped binaries (debug symbols excluded)
Baseline Size: 37.08 MiB
Comparison Size: 37.08 MiB
Size Change: -8 B (-0.00%)
Pass/Fail Threshold: +5%
Result: PASSED ✅

Changes by Module

Module	File Size	Symbols
`anon.4f8fd67d74ae1f1600187cfeb0121be9.1.llvm.14393392841373718055`	-130 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.1.llvm.6868297941858885019`	+129 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.4.llvm.14393392841373718055`	-115 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.4.llvm.6868297941858885019`	+114 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.3.llvm.14393392841373718055`	-109 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.3.llvm.6868297941858885019`	+108 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.0.llvm.14393392841373718055`	-97 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.0.llvm.6868297941858885019`	+96 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.2.llvm.14393392841373718055`	-95 B	1
`anon.4f8fd67d74ae1f1600187cfeb0121be9.2.llvm.6868297941858885019`	+94 B	1
`[Unmapped]`	-3 B	1

Detailed Symbol Changes

    FILE SIZE        VM SIZE    
 --------------  -------------- 
  [NEW]    +129  [NEW]     +40    anon.4f8fd67d74ae1f1600187cfeb0121be9.1.llvm.6868297941858885019
  [NEW]    +114  [NEW]     +25    anon.4f8fd67d74ae1f1600187cfeb0121be9.4.llvm.6868297941858885019
  [NEW]    +108  [NEW]     +19    anon.4f8fd67d74ae1f1600187cfeb0121be9.3.llvm.6868297941858885019
  [NEW]     +96  [NEW]      +7    anon.4f8fd67d74ae1f1600187cfeb0121be9.0.llvm.6868297941858885019
  [NEW]     +94  [NEW]      +5    anon.4f8fd67d74ae1f1600187cfeb0121be9.2.llvm.6868297941858885019
  -0.1%      -3  [ = ]       0    [Unmapped]
  [DEL]     -95  [DEL]      -5    anon.4f8fd67d74ae1f1600187cfeb0121be9.2.llvm.14393392841373718055
  [DEL]     -97  [DEL]      -7    anon.4f8fd67d74ae1f1600187cfeb0121be9.0.llvm.14393392841373718055
  [DEL]    -109  [DEL]     -19    anon.4f8fd67d74ae1f1600187cfeb0121be9.3.llvm.14393392841373718055
  [DEL]    -115  [DEL]     -25    anon.4f8fd67d74ae1f1600187cfeb0121be9.4.llvm.14393392841373718055
  [DEL]    -130  [DEL]     -40    anon.4f8fd67d74ae1f1600187cfeb0121be9.1.llvm.14393392841373718055
  -0.0%      -8  [ = ]       0    TOTAL

pr-commenter · 2026-05-05T11:46:22Z

Regression Detector (Agent Data Plane)

Regression Detector Results

Run ID: f7284ee9-1a8f-4882-bcae-e2092f84ef08

Baseline: e8a875d
Comparison: 803ddec
Diff

Optimization Goals: ✅ No significant changes detected

Experiments ignored for regressions

Regressions in experiments with settings containing erratic: true are ignored.

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
❌	otlp_ingest_logs_5mb_memory	memory utilization	+10.25	[+9.78, +10.71]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_logs_5mb_cpu	% cpu utilization	+1.38	[-3.53, +6.28]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_logs_5mb_throughput	ingress throughput	-0.02	[-0.15, +0.10]	1	(metrics) (profiles) (logs)

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
❌	otlp_ingest_logs_5mb_memory	memory utilization	+10.25	[+9.78, +10.71]	1	(metrics) (profiles) (logs)
➖	dsd_uds_512kb_3k_contexts_cpu	% cpu utilization	+7.64	[-50.37, +65.65]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_logs_5mb_cpu	% cpu utilization	+1.38	[-3.53, +6.28]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_metrics_5mb_cpu	% cpu utilization	+0.44	[-5.69, +6.57]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_ottl_filtering_5mb_memory	memory utilization	+0.34	[+0.10, +0.57]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_ottl_transform_5mb_cpu	% cpu utilization	+0.25	[-1.96, +2.45]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_5mb_throughput	ingress throughput	+0.17	[+0.09, +0.24]	1	(metrics) (profiles) (logs)
➖	quality_gates_rss_dsd_low	memory utilization	+0.16	[+0.00, +0.32]	1	(metrics) (profiles) (logs)
➖	dsd_uds_500mb_3k_contexts_throughput	ingress throughput	+0.06	[-0.06, +0.19]	1	(metrics) (profiles) (logs)
➖	quality_gates_rss_dsd_ultraheavy	memory utilization	+0.06	[-0.08, +0.20]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_ottl_transform_5mb_memory	memory utilization	+0.06	[-0.10, +0.22]	1	(metrics) (profiles) (logs)
➖	quality_gates_rss_dsd_medium	memory utilization	+0.06	[-0.11, +0.23]	1	(metrics) (profiles) (logs)
➖	dsd_uds_10mb_3k_contexts_throughput	ingress throughput	+0.02	[-0.18, +0.21]	1	(metrics) (profiles) (logs)
➖	dsd_uds_512kb_3k_contexts_throughput	ingress throughput	+0.01	[-0.05, +0.06]	1	(metrics) (profiles) (logs)
➖	dsd_uds_1mb_3k_contexts_throughput	ingress throughput	-0.00	[-0.06, +0.05]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_metrics_5mb_throughput	ingress throughput	-0.01	[-0.14, +0.12]	1	(metrics) (profiles) (logs)
➖	dsd_uds_100mb_3k_contexts_throughput	ingress throughput	-0.01	[-0.06, +0.03]	1	(metrics) (profiles) (logs)
➖	dsd_uds_10mb_3k_contexts_memory	memory utilization	-0.02	[-0.18, +0.14]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_logs_5mb_throughput	ingress throughput	-0.02	[-0.15, +0.10]	1	(metrics) (profiles) (logs)
➖	dsd_uds_512kb_3k_contexts_memory	memory utilization	-0.03	[-0.17, +0.12]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_5mb_memory	memory utilization	-0.03	[-0.19, +0.13]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_ottl_transform_5mb_throughput	ingress throughput	-0.07	[-0.15, -0.00]	1	(metrics) (profiles) (logs)
➖	quality_gates_rss_dsd_heavy	memory utilization	-0.09	[-0.22, +0.04]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_ottl_filtering_5mb_throughput	ingress throughput	-0.10	[-0.17, -0.02]	1	(metrics) (profiles) (logs)
➖	dsd_uds_100mb_3k_contexts_memory	memory utilization	-0.10	[-0.25, +0.05]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_5mb_cpu	% cpu utilization	-0.14	[-2.19, +1.91]	1	(metrics) (profiles) (logs)
➖	dsd_uds_1mb_3k_contexts_memory	memory utilization	-0.17	[-0.30, -0.03]	1	(metrics) (profiles) (logs)
➖	dsd_uds_500mb_3k_contexts_memory	memory utilization	-0.20	[-0.35, -0.06]	1	(metrics) (profiles) (logs)
➖	dsd_uds_500mb_3k_contexts_cpu	% cpu utilization	-0.49	[-1.94, +0.96]	1	(metrics) (profiles) (logs)
➖	quality_gates_rss_idle	memory utilization	-0.56	[-0.60, -0.51]	1	(metrics) (profiles) (logs)
➖	dsd_uds_100mb_3k_contexts_cpu	% cpu utilization	-1.43	[-7.47, +4.61]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_metrics_5mb_memory	memory utilization	-1.53	[-1.75, -1.32]	1	(metrics) (profiles) (logs)
➖	otlp_ingest_traces_ottl_filtering_5mb_cpu	% cpu utilization	-2.35	[-4.50, -0.20]	1	(metrics) (profiles) (logs)
➖	dsd_uds_10mb_3k_contexts_cpu	% cpu utilization	-2.83	[-31.76, +26.10]	1	(metrics) (profiles) (logs)
➖	dsd_uds_1mb_3k_contexts_cpu	% cpu utilization	-3.08	[-54.90, +48.73]	1	(metrics) (profiles) (logs)

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	quality_gates_rss_dsd_heavy	memory_usage	10/10	123.57MiB ≤ 140MiB	(metrics) (profiles) (logs)
✅	quality_gates_rss_dsd_low	memory_usage	10/10	40.36MiB ≤ 50MiB	(metrics) (profiles) (logs)
✅	quality_gates_rss_dsd_medium	memory_usage	10/10	60.71MiB ≤ 75MiB	(metrics) (profiles) (logs)
✅	quality_gates_rss_dsd_ultraheavy	memory_usage	10/10	174.64MiB ≤ 200MiB	(metrics) (profiles) (logs)
✅	quality_gates_rss_idle	memory_usage	10/10	27.59MiB ≤ 40MiB	(metrics) (profiles) (logs)

Explanation

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

Performance changes are noted in the perf column of each table:

✅ = significantly better comparison variant performance
❌ = significantly worse comparison variant performance
➖ = no significant change in performance

A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".

For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:

Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.
Its configuration does not mark it "erratic".

tobz

Mostly non-blocking nits.

tobz · 2026-05-05T12:43:49Z

+
+Level 4: At the user's discretion, proceed to integration testing. There are two integration test
+suites:
+- Name "integration": test definitions at `test/integration`.


Sidenote: we should probably move the integration test cases into an actual subdirectory called cases to match all the other test groupings we have.

Agreed. Not in this one though. I also want to move some panoramic code for clearer patterns.

@echo

## Summary Claude was routinely opting not to read AGENTS.md. I found that even when I added a `.claude/CLAUDE.md` which tells it to read AGENTS.md it still skipped that step. I added a session start hook that more-or-less forces Claude to actually read the AGENTS.md file and I added a section about our building and testing commands to AGENTS.md I also added some information to help it understand what commands we need to use to check our work. It was routinely reporting being done with work without running fmt or the linter. I think it's still going to skip those steps but at least now we can scold it and say "Did you run the linting steps?" and it will have those instructions available to say "Oh sorry, blah blah, here i go with `make fmt`" Note: some of it could get stale quickly. We should actually create a skill and just run it occasionally to look for stale instructions or paths. But keeping this PR quick and small. ## Change Type - [x] Non-functional (chore, refactoring, docs) ## How did you test this PR? ### Before On main, when I start a session and say "run the integration tests", it looks around and chooses `make test-integration-quick` to save time, i.e. not building the containers. That's most likely not what you want: ``` ⏺ Let me check how integration tests are run in this project. ⏺ Bash(ls /my/weird/path//saluki-main/) ⎿ AGENTS.md bin bun.lock Cargo.lock Cargo.toml ci clippy.toml deny.toml docker docs lib LICENSE LICENSE-3rdparty.csv license-tool.toml Makefile NOTICE package.json README.md renovate.json5 repository.datadog.yml rust-toolchain.toml rustfmt.toml SECURITY.md static-analysis.datadog.yml target test tomlfmt.toml tooling ⏺ Bash(grep -i "integrat\|test-correct\|test-integration" /my/weird/path//saluki-main/Makefile | head -40) ⎿ .PHONY: test-correctness test-correctness: build-panoramic test-correctness: ## Runs the complete correctness suite (all test cases in parallel) .PHONY: test-correctness-case test-correctness-case: build-panoramic test-correctness-case: ## Runs a single correctness test case by name (usage: make test-correctness-case CASE=dsd-plain) build-panoramic: ## Builds the panoramic binary (ADP integration test runner) .PHONY: test-integration test-integration: build-panoramic build-datadog-agent-image test-integration: ## Runs all ADP integration tests @echo "[*] Running ADP integration tests..." @target/release/panoramic run -d $(shell pwd)/test/integration/cases $(if $(PANORAMIC_LOG_DIR),-l $(PANORAMIC_LOG_DIR)) .PHONY: test-integration-quick test-integration-quick: build-panoramic test-integration-quick: ## Runs ADP integration tests (assumes images already built) @echo "[*] Running ADP integration tests (quick mode)..." @target/release/panoramic run -d $(shell pwd)/test/integration/cases $(if $(PANORAMIC_LOG_DIR),-l $(PANORAMIC_LOG_DIR)) .PHONY: list-integration-tests list-integration-tests: build-panoramic list-integration-tests: ## Lists available ADP integration tests @target/release/panoramic list -d $(shell pwd)/test/integration/cases ⏺ Bash(make test-integration-quick 2>&1) ``` ### After It reads `AGENTS.md` and understands what to do: ``` ⏺ Let me read AGENTS.md first as instructed. Searched for 1 pattern, read 1 file (ctrl+o to expand) ⏺ Let me check the Makefile to confirm the integration test targets. Searched for 1 pattern, read 1 file (ctrl+o to expand) ⏺ The integration suite uses make test-integration, which builds images first. Running it now — this will take a while since it builds the Datadog Agent image. ⏺ Bash(make test-integration 2>&1) ⎿ Running in the background (↓ to manage) ⎿ (timeout 10m) ``` ## References N/A 7223534

chore(docs): add ai build and test instructions

875c30a

Add build and test instructions to AGENTS.md to improve Claude Code performance.

webern requested a review from a team as a code owner May 5, 2026 11:33

chatgpt-codex-connector Bot reviewed May 5, 2026

View reviewed changes

Comment thread AGENTS.md Outdated

webern commented May 5, 2026

View reviewed changes

Comment thread AGENTS.md Outdated

webern added 2 commits May 5, 2026 13:36

Update AGENTS.md

460ec87

fix spelling because i do not know how to make codex fix it

e2b8867

tobz approved these changes May 5, 2026

View reviewed changes

tobz added the type/chore Updates to dependencies or general "administrative" tasks necessary to maintain the codebase/repo. label May 5, 2026

fixup: pr suggestions

803ddec

webern merged commit 7223534 into main May 5, 2026
72 checks passed

webern deleted the matt.briggs/agents.md branch May 5, 2026 16:25

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

chore(docs): improve ai instructions#1582

chore(docs): improve ai instructions#1582
webern merged 4 commits into
mainfrom
matt.briggs/agents.md

webern commented May 5, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

Uh oh!

Uh oh!

pr-commenter Bot commented May 5, 2026 •

edited

Loading

Changes by Module

Detailed Symbol Changes

Uh oh!

pr-commenter Bot commented May 5, 2026 •

edited

Loading

Experiments ignored for regressions

Fine details of change detection per experiment

Bounds Checks: ✅ Passed

Explanation

Uh oh!

tobz left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tobz May 5, 2026

Uh oh!

webern May 5, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

webern commented May 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Change Type

How did you test this PR?

Before

After

References

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

pr-commenter Bot commented May 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Binary Size Analysis (Agent Data Plane)

Changes by Module

Detailed Symbol Changes

Uh oh!

pr-commenter Bot commented May 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Regression Detector (Agent Data Plane)

Regression Detector Results

Optimization Goals: ✅ No significant changes detected

Experiments ignored for regressions

Fine details of change detection per experiment

Bounds Checks: ✅ Passed

Explanation

Uh oh!

tobz left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tobz May 5, 2026

Choose a reason for hiding this comment

Uh oh!

webern May 5, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

webern commented May 5, 2026 •

edited

Loading

pr-commenter Bot commented May 5, 2026 •

edited

Loading

pr-commenter Bot commented May 5, 2026 •

edited

Loading