Join our Newsletter — 33% off our NHI Course
Home› Guides› Secrets Management Executive Briefing
Executive Briefing Secrets Management

Secrets Management Executive Briefing

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 14 November 2026 10 min read
On this page

The passwords, API keys, tokens and certificates that let your software talk to other systems are among the most valuable things an attacker can steal, and in most organisations they are scattered across code, configuration files, build pipelines and chat messages. This briefing is for chief executives, CFOs, CIOs, risk and audit committee members and the CISOs who report to them. It explains secrets risk in business terms, what a secrets management programme buys, the questions that test whether management has it under control, where to invest and a 90-day plan. The wider question of who owns and governs machine identities is covered in the Board Briefing: NHI Risk; this page concentrates on the credentials themselves and how they are stored, delivered, rotated and revoked.

The risk in one paragraph

A secret is a credential used by software rather than a person: a database password in a configuration file, a cloud access key in a build pipeline, an API token for a payment provider, a private key for a certificate. Anyone who holds a copy can use it, usually without multi-factor authentication and often from anywhere on the internet. Developers create secrets every day, copy them to wherever they are needed and rarely retire them, so each one tends to outlive the project that created it and to exist in several places at once. When one leaks, through a public code repository, a log file, a shared document or a compromised laptop, the attacker gets trusted access that looks like normal system traffic. Three things set the size of the loss, and management can report on all three: how many copies of each secret exist outside a controlled store, how much access each one grants, and how long it stays valid after it is exposed.

Why this is a board-level issue now

  • Leaks happen at industrial scale and stay usable. GitGuardian's State of Secrets Sprawl 2025 report counted 23,770,171 new hard-coded secrets on public GitHub in 2024, up 25 per cent on the previous year, and found that 70 per cent of secrets leaked in 2022 were still valid. Finding a leak is not the same as fixing it.
  • Private code is not safe code. The same report found plaintext secrets in 35 per cent of the private repositories it scanned for customers. A secret in a private repository is exposed to everyone with access to that repository, every fork and every laptop that clones it, and to any attacker who compromises one of them.
  • One exposed token can stay open for a year. In the Home Depot token exposure, a GitHub access token published by an employee was reported as exposed from early 2024 until December 2025, giving access to private source code, cloud infrastructure and order fulfilment systems. It was revoked only after a journalist contacted the company.
  • Controls only work if nobody can switch them off. In the CISA Private-CISA GitHub leak, a contractor had disabled the protection that blocks secrets from being pushed to public repositories, and administrative cloud keys stayed valid for about 48 hours after the repository was taken down.
  • Directors are expected to oversee this kind of risk. The UK Cyber Governance Code of Practice (DSIT, 8 April 2025) asks boards to agree a cyber risk appetite (A3), confirm that resources match the agreed risks (B3) and require formal reporting at least quarterly against agreed metrics and tolerances (E2). Secrets exposure is measurable and fits that reporting directly.

What a secrets management programme actually does

Boards often hear "we have a vault" and stop there. A secrets manager is a tool; the programme around it is what reduces risk. The OWASP Secrets Management Cheat Sheet describes the practice in five parts that translate cleanly into business outcomes.

Part of the programmeWhat it means in practiceBusiness outcome
CentraliseProduction secrets live in an approved store with access control and audit logs, not in code, files or chatManagement knows where secrets are and who can read them
PreventScanning in developer tools and push protection on code repositories block new secrets from being committedFewer leaks to clean up, lower incident volume
Shorten lifetimeSecrets expire and are rotated automatically; where possible, short-lived credentials are issued on demand instead of storedA stolen secret is worth hours, not years
MonitorEvery read of a secret is logged, and unusual access raises an alertMisuse is noticed rather than discovered by a third party
RespondA playbook to revoke, rotate, remove and review, in that orderExposure windows measured in minutes

The most important shift for leadership to understand is the last step of maturity: replacing stored secrets altogether. Modern cloud platforms and build systems can prove a workload's identity to another service and receive a short-lived credential each time, a pattern usually called workload identity federation. The OWASP cheat sheet recommends short-lived OIDC tokens and federation for build pipelines instead of static keys. Every key removed this way is one fewer thing to store, rotate or leak. See Static vs Dynamic Secrets.

Five questions for the board to ask

  1. What share of our production secrets is held in an approved secrets manager, and how is that measured? A credible answer is a percentage from scanning code, pipelines and cloud configuration, not "all new applications use the vault".
  2. If a key to our main cloud account appeared on the internet tonight, how long would it take to make it useless? Expect a time measured in previous incidents or exercises, the names of the people allowed to revoke it out of hours and confirmation that revocation would not stop the business.
  3. Can anyone in the company switch off the controls that stop secrets being committed to code? The answer should name a small group allowed to approve exceptions, and every bypass should be logged and reviewed.
  4. How old is our oldest production credential, and which of our critical systems still depend on secrets that never expire? A good answer gives the age distribution and a dated plan for the oldest, highest-privilege secrets.
  5. Who outside the company holds our secrets, and how would we know if they leaked them? Contractors, managed service providers and SaaS integrations all hold credentials into your systems; management should know which, with what access, and how they are rotated when a contract ends.

Red flags in management answers

  • "We bought a vault": tool ownership reported instead of coverage.
  • "We scan for secrets": scanning that finds leaks but has no owner or deadline to revoke them.
  • "Rotation would break things": a sign that the dependencies on each secret are not documented.
  • "It is in a private repository": private code treated as a safe place for credentials.

What good looks like

AreaWeak positionGood position
Where secrets liveCode, .env files, pipeline variables, shared documentsApproved store for all production secrets, with an owner and purpose recorded for each
New leaksFound by researchers, journalists or attackersBlocked at commit; bypasses limited and reviewed
LifetimeKeys valid for yearsMaximum lifetimes by secret type; short-lived credentials for pipelines and cloud workloads
ResponseFile deleted, secret left validRevocation first, rotation through an automated process, review of access logs over the exposure window
Third partiesShared keys with no expiryScoped credentials per supplier, rotated at contract changes
ReportingNo figuresQuarterly metrics with agreed tolerances, as the Cyber Governance Code expects

To score the current position in detail, management can use the Secrets Management Maturity Model.

Where to invest first

  1. Find what is already exposed: scan code history, build logs, container images and collaboration tools, then revoke and rotate the highest-privilege findings. This is usually the fastest risk reduction available. See the Secret Sprawl Measurement Guide.
  2. Stop new leaks: push protection and pre-commit scanning on every repository, with bypass limited to a delegated group.
  3. Centralise production secrets: move them into an approved store, starting with cloud administrator keys, database credentials and payment or customer data integrations.
  4. Remove secrets from build pipelines: replace stored cloud keys in continuous integration with federated, short-lived credentials.
  5. Rehearse a leak: a timed exercise from "key found on the internet" to "key useless and access reviewed", including out of hours.

Most of the cost is engineering time to migrate applications and change habits, not licences. For selecting tooling, see the Secrets Management Buyer's Guide.

Metrics to track

  • Share of production secrets held in an approved store, by business unit.
  • Number of valid secrets found outside approved stores, and the trend quarter on quarter.
  • Median and longest time from a secret leak being reported to the secret being revoked.
  • Share of high-privilege secrets older than the agreed maximum lifetime. See How to measure secret age and rotation coverage.
  • Share of build pipelines that deploy to cloud using federated credentials rather than stored keys.
  • Number of push protection bypasses, and the share reviewed within the agreed time.

A 90-day plan

  • Days 1 to 30: name an executive owner for secrets risk; ask for a baseline from scanning code, pipelines and cloud configuration; get the list of the ten highest-privilege secrets and their age; confirm who can revoke a leaked secret out of hours.
  • Days 31 to 60: revoke and rotate the most serious exposed secrets; turn on push protection across all repositories with delegated bypass; agree maximum lifetimes by secret type; adopt a written leak response playbook.
  • Days 61 to 90: run a timed leak exercise; set targets for store coverage and federated pipelines; agree board tolerances for the metrics above and the first quarterly report.

Common failure patterns

  • Deleting instead of revoking: the file or commit is removed but the secret still works, and copies remain in forks, caches and logs.
  • Coverage claimed from new projects: new applications use the store while older systems, which hold the most powerful secrets, are left out.
  • Controls with an off switch: push protection or scanning that any developer can bypass or disable without review.
  • Rotation without dependency mapping: rotation avoided because nobody knows what would break, so secrets stay valid for years.
  • The vault as a single point of trust: the secrets manager's own administrator credentials protected less well than the secrets it holds. See If your secrets manager is compromised.

More incidents and their lessons are collected in Leaked secret breaches.

Framework anchors for the audit committee

FrameworkReferenceWhat it expects
NIST SP 800-53 Rev. 5IA-5, IA-5(7)Authenticator management; unencrypted static authenticators not embedded in applications or other static storage
NIST SP 800-53 Rev. 5SC-12Cryptographic key establishment and management
NIST CSF 2.0PR.AA-01; GV.RM-03Credentials for services managed; cyber risk included in enterprise risk management
OWASP Non-Human Identities Top 10 (2025)NHI2 Secret Leakage; NHI7 Long-Lived SecretsThe two risks this programme most directly reduces
MITRE ATT&CK v20T1552.001 Credentials In FilesThe attacker technique that hard-coded secrets enable
UK Cyber Governance Code of PracticeA3, B3, E2Risk appetite, resourcing and at least quarterly reporting with metrics and tolerances

Standards and references

How NHI Mgmt Group can help

We provide independent secrets and non-human identity risk assessments, board briefings and programme initiation support. Our NHI Foundation Level Training Course helps security, platform and IAM teams govern service accounts, API keys and secrets. Contact us to arrange a briefing.

Related NHI Mgmt Group resources: Secrets Management Guide · Secrets Management Maturity Model · Board Briefing: NHI Risk · Secrets Management Health Check

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 14 November 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.